Pushmeet Kohli

TMLR Journal 2025 Journal Article

Privacy Awareness for Information-Sharing Assistants: A Case-study on Form-filling with Contextual Integrity

• Sahra Ghalebikesabi
• Eugene Bagdasarian
• Ren Yi
• Itay Yona
• Ilia Shumailov
• Aneesh Pappu
• Chongyang Shi
• Laura Weidinger

Advanced AI assistants combine frontier LLMs and tool access to autonomously perform complex tasks on behalf of users. While the helpfulness of such assistants can increase dramatically with access to user information including emails and documents, this raises privacy concerns about assistants sharing inappropriate information with third parties without user supervision. To steer information-sharing assistants to behave in accordance with privacy expectations, we propose to operationalize the design of privacy-conscious assistants that conform with *contextual integrity* (CI), a framework that equates privacy with the appropriate flow of information in a given context. In particular, we design and evaluate a number of strategies to steer assistants' information-sharing actions to be CI compliant. Our evaluation is based on a novel form filling benchmark composed of human annotations of common webform applications, and it reveals that prompting frontier LLMs to perform CI-based reasoning yields strong results.

ICLR Conference 2025 Conference Paper

Scaling Wearable Foundation Models

• Girish Narayanswamy
• Xin Liu 0034
• Kumar Ayush
• Yuzhe Yang 0003
• Xuhai Xu
• Shun Liao
• Jake Garrison
• Shyam A. Tailor

Wearable sensors have become ubiquitous thanks to a variety of health tracking features. The resulting continuous and longitudinal measurements from everyday life generate large volumes of data. However, making sense of these observations for scientific and actionable insights is non-trivial. Inspired by the empirical success of generative modeling, where large neural networks learn powerful representations from vast amounts of text, image, video, or audio data, we investigate the scaling properties of wearable sensor foundation models across compute, data, and model size. Using a dataset of up to 40 million hours of in-situ heart rate, heart rate variability, accelerometer, electrodermal activity, skin temperature, and altimeter per-minute data from over 165,000 people, we create LSM, a multimodal foundation model built on the largest wearable-signals dataset with the most extensive range of sensor modalities to date. Our results establish the scaling laws of LSM for tasks such as imputation, interpolation and extrapolation across both time and sensor modalities. Moreover, we highlight how LSM enables sample-efficient downstream learning for tasks including exercise and activity recognition.

NeurIPS Conference 2025 Conference Paper

SensorLM: Learning the Language of Wearable Sensors

• Yuwei Zhang
• Kumar Ayush
• Siyuan Qiao
• A. Ali Heydari
• Girish Narayanswamy
• Max Xu
• Ahmed Metwally
• Jinhua Xu

We present SensorLM, a family of sensor-language foundation models that enable wearable sensor data understanding with natural language. Despite its pervasive nature, aligning and interpreting sensor data with language remains challenging due to the lack of paired, richly annotated sensor-text descriptions in uncurated, real-world wearable data. We introduce a hierarchical caption generation pipeline designed to capture statistical, structural, and semantic information from sensor data. This approach enabled the curation of the largest sensor-language dataset to date, comprising over 59. 7 million hours of data from more than 103, 000 people. Furthermore, SensorLM extends prominent multimodal pretraining architectures (e. g. , CLIP, CoCa) and recovers them as specific variants within a generic architecture. Extensive experiments on real-world tasks in human activity analysis and healthcare verify the superior performance of SensorLM over state-of-the-art in zero-shot recognition, few-shot learning, and cross-modal retrieval. SensorLM also demonstrates intriguing capabilities including scaling behaviors, label efficiency, sensor captioning, and zero-shot generalization to unseen tasks. Code is available at https: //github. com/Google-Health/consumer-health-research/tree/main/sensorlm.

IJCAI Conference 2022 Conference Paper

Making Sense of Raw Input (Extended Abstract)

• Richard Evans
• Matko Bošnjak
• Lars Buesing
• Kevin Ellis
• David Pfau
• Pushmeet Kohli
• Marek Sergot

How should a machine intelligence perform unsupervised structure discovery over streams of sensory input? One approach to this problem is to cast it as an apperception task. Here, the task is to construct an explicit interpretable theory that both explains the sensory sequence and also satisfies a set of unity conditions, designed to ensure that the constituents of the theory are connected in a relational structure. However, the original formulation of the apperception task had one fundamental limitation: it assumed the raw sensory input had already been parsed using a set of discrete categories, so that all the system had to do was receive this already-digested symbolic input, and make sense of it. But what if we don't have access to pre-parsed input? What if our sensory sequence is raw unprocessed information? The central contribution of this paper is a neuro-symbolic framework for distilling interpretable theories out of streams of raw, unprocessed sensory experience. First, we extend the definition of the apperception task to include ambiguous (but still symbolic) input: sequences of sets of disjunctions. Next, we use a neural network to map raw sensory input to disjunctive input. Our binary neural network is encoded as a logic program, so the weights of the network and the rules of the theory can be solved jointly as a single SAT problem. This way, we are able to jointly learn how to perceive (mapping raw sensory information to concepts) and apperceive (combining concepts into declarative rules).

AIJ Journal 2021 Journal Article

Making sense of raw input

• Richard Evans
• Matko Bošnjak
• Lars Buesing
• Kevin Ellis
• David Pfau
• Pushmeet Kohli
• Marek Sergot

How should a machine intelligence perform unsupervised structure discovery over streams of sensory input? One approach to this problem is to cast it as an apperception task [1]. Here, the task is to construct an explicit interpretable theory that both explains the sensory sequence and also satisfies a set of unity conditions, designed to ensure that the constituents of the theory are connected in a relational structure. However, the original formulation of the apperception task had one fundamental limitation: it assumed the raw sensory input had already been parsed using a set of discrete categories, so that all the system had to do was receive this already-digested symbolic input, and make sense of it. But what if we don't have access to pre-parsed input? What if our sensory sequence is raw unprocessed information? The central contribution of this paper is a neuro-symbolic framework for distilling interpretable theories out of streams of raw, unprocessed sensory experience. First, we extend the definition of the apperception task to include ambiguous (but still symbolic) input: sequences of sets of disjunctions. Next, we use a neural network to map raw sensory input to disjunctive input. Our binary neural network is encoded as a logic program, so the weights of the network and the rules of the theory can be solved jointly as a single SAT problem. This way, we are able to jointly learn how to perceive (mapping raw sensory information to concepts) and apperceive (combining concepts into declarative rules).

AIJ Journal 2021 Journal Article

Making sense of sensory input

• Richard Evans
• José Hernández-Orallo
• Johannes Welbl
• Pushmeet Kohli
• Marek Sergot

This paper attempts to answer a central question in unsupervised learning: what does it mean to “make sense” of a sensory sequence? In our formalization, making sense involves constructing a symbolic causal theory that both explains the sensory sequence and also satisfies a set of unity conditions. The unity conditions insist that the constituents of the causal theory – objects, properties, and laws – must be integrated into a coherent whole. On our account, making sense of sensory input is a type of program synthesis, but it is unsupervised program synthesis. Our second contribution is a computer implementation, the Apperception Engine, that was designed to satisfy the above requirements. Our system is able to produce interpretable human-readable causal theories from very small amounts of data, because of the strong inductive bias provided by the unity conditions. A causal theory produced by our system is able to predict future sensor readings, as well as retrodict earlier readings, and impute (fill in the blanks of) missing sensory readings, in any combination. In fact, it is able to do all three tasks simultaneously. We tested the engine in a diverse variety of domains, including cellular automata, rhythms and simple nursery tunes, multi-modal binding problems, occlusion tasks, and sequence induction intelligence tests. In each domain, we test our engine's ability to predict future sensor values, retrodict earlier sensor values, and impute missing sensory data. The Apperception Engine performs well in all these domains, significantly out-performing neural net baselines. We note in particular that in the sequence induction intelligence tests, our system achieved human-level performance. This is notable because our system is not a bespoke system designed specifically to solve intelligence tests, but a general-purpose system that was designed to make sense of any sensory sequence.

ICLR Conference 2021 Conference Paper

Self-supervised Adversarial Robustness for the Low-label, High-data Regime

• Sven Gowal
• Po-Sen Huang
• Aäron van den Oord
• Timothy A. Mann
• Pushmeet Kohli

Recent work discovered that training models to be invariant to adversarial perturbations requires substantially larger datasets than those required for standard classification. Perhaps more surprisingly, these larger datasets can be "mostly" unlabeled. Pseudo-labeling, a technique simultaneously pioneered by four separate and simultaneous works in 2019, has been proposed as a competitive alternative to labeled data for training adversarially robust models. However, when the amount of labeled data decreases, the performance of pseudo-labeling catastrophically drops, thus questioning the theoretical insights put forward by Uesato et al. (2019), which suggest that the sample complexity for learning an adversarially robust model from unlabeled data should match the fully supervised case. We introduce Bootstrap Your Own Robust Latents (BYORL), a self-supervised learning technique based on BYOL for training adversarially robust models. Our method enables us to train robust representations without any labels (reconciling practice with theory). Most notably, this robust representation can be leveraged by a linear classifier to train adversarially robust models, even when the linear classifier is not trained adversarially. We evaluate BYORL and pseudo-labeling on CIFAR-10 and ImageNet and demonstrate that BYORL achieves significantly higher robustness (i.e., models resulting from BYORL are up to two times more accurate). Experiments on CIFAR-10 against $\ell_2$ and $\ell_\infty$ norm-bounded perturbations demonstrate that BYORL achieves near state-of-the-art robustness with as little as 500 labeled examples. We also note that against $\ell_2$ norm-bounded perturbations of size $\epsilon = 128/255$, BYORL surpasses the known state-of-the-art with an accuracy under attack of 77.61% (against 72.91% for the prior art).

ICLR Conference 2020 Conference Paper

A Framework for robustness Certification of Smoothed Classifiers using F-Divergences

• Krishnamurthy Dvijotham
• Jamie Hayes
• Borja Balle
• J. Zico Kolter
• Chongli Qin
• András György 0001
• Kai Xiao
• Sven Gowal

Formal verification techniques that compute provable guarantees on properties of machine learning models, like robustness to norm-bounded adversarial perturbations, have yielded impressive results. Although most techniques developed so far require knowledge of the architecture of the machine learning model and remain hard to scale to complex prediction pipelines, the method of randomized smoothing has been shown to overcome many of these obstacles. By requiring only black-box access to the underlying model, randomized smoothing scales to large architectures and is agnostic to the internals of the network. However, past work on randomized smoothing has focused on restricted classes of smoothing measures or perturbations (like Gaussian or discrete) and has only been able to prove robustness with respect to simple norm bounds. In this paper we introduce a general framework for proving robustness properties of smoothed machine learning models in the black-box setting. Specifically, we extend randomized smoothing procedures to handle arbitrary smoothing measures and prove robustness of the smoothed classifier by using f-divergences. Our methodology improves upon the state of the art in terms of computation time or certified robustness on several image classification tasks and an audio classification task, with respect to several classes of adversarial perturbations.

ICLR Conference 2020 Conference Paper

Adversarially Robust Representations with Smooth Encoders

• A. Taylan Cemgil
• Sumedh Ghaisas
• Krishnamurthy Dvijotham
• Pushmeet Kohli

This paper studies the undesired phenomena of over-sensitivity of representations learned by deep networks to semantically-irrelevant changes in data. We identify a cause for this shortcoming in the classical Variational Auto-encoder (VAE) objective, the evidence lower bound (ELBO). We show that the ELBO fails to control the behaviour of the encoder out of the support of the empirical data distribution and this behaviour of the VAE can lead to extreme errors in the learned representation. This is a key hurdle in the effective use of representations for data-efficient learning and transfer. To address this problem, we propose to augment the data with specifications that enforce insensitivity of the representation with respect to families of transformations. To incorporate these specifications, we propose a regularization method that is based on a selection mechanism that creates a fictive data point by explicitly perturbing an observed true data point. For certain choices of parameters, our formulation naturally leads to the minimization of the entropy regularized Wasserstein distance between representations. We illustrate our approach on standard datasets and experimentally show that significant improvements in the downstream adversarial accuracy can be achieved by learning robust representations completely in an unsupervised manner, without a reference to a particular downstream task and without a costly supervised adversarial training procedure.

JMLR Journal 2020 Journal Article

Branch and Bound for Piecewise Linear Neural Network Verification

• Rudy Bunel
• Jingyue Lu
• Ilker Turkaslan
• Philip H.S. Torr
• Pushmeet Kohli
• M. Pawan Kumar

The success of Deep Learning and its potential use in many safety-critical applicationshas motivated research on formal verification of Neural Network (NN) models. In thiscontext, verification involves proving or disproving that an NN model satisfies certaininput-output properties. Despite the reputation of learned NN models as black boxes,and the theoretical hardness of proving useful properties about them, researchers havebeen successful in verifying some classes of models by exploiting their piecewise linearstructure and taking insights from formal methods such as Satisifiability Modulo Theory.However, these methods are still far from scaling to realistic neural networks. To facilitateprogress on this crucial area, we exploit the Mixed Integer Linear Programming (MIP) formulation of verification to propose a family of algorithms based on Branch-and-Bound (BaB). We show that our family contains previous verification methods as special cases.With the help of the BaB framework, we make three key contributions. Firstly, we identifynew methods that combine the strengths of multiple existing approaches, accomplishingsignificant performance improvements over previous state of the art. Secondly, we introducean effective branching strategy on ReLU non-linearities. This branching strategy allows usto efficiently and successfully deal with high input dimensional problems with convolutionalnetwork architecture, on which previous methods fail frequently. Finally, we proposecomprehensive test data sets and benchmarks which includes a collection of previouslyreleased testcases. We use the data sets to conduct a thorough experimental comparison ofexisting and new algorithms and to provide an inclusive analysis of the factors impactingthe hardness of verification problems. [abs] [ pdf ][ bib ] &copy JMLR 2020. ( edit, beta )

ICLR Conference 2020 Conference Paper

CLEVRER: Collision Events for Video Representation and Reasoning

• Kexin Yi
• Chuang Gan 0001
• Yunzhu Li
• Pushmeet Kohli
• Jiajun Wu 0001
• Antonio Torralba 0001
• Joshua B. Tenenbaum

The ability to reason about temporal and causal events from videos lies at the core of human intelligence. Most video reasoning benchmarks, however, focus on pattern recognition from complex visual and language input, instead of on causal structure. We study the complementary problem, exploring the temporal and causal structures behind videos of objects with simple visual appearance. To this end, we introduce the CoLlision Events for Video REpresentation and Reasoning (CLEVRER) dataset, a diagnostic video dataset for systematic evaluation of computational models on a wide range of reasoning tasks. Motivated by the theory of human casual judgment, CLEVRER includes four types of question: descriptive (e.g., ‘what color’), explanatory (‘what’s responsible for’), predictive (‘what will happen next’), and counterfactual (‘what if’). We evaluate various state-of-the-art models for visual reasoning on our benchmark. While these models thrive on the perception-based task (descriptive), they perform poorly on the causal tasks (explanatory, predictive and counterfactual), suggesting that a principled approach for causal reasoning should incorporate the capability of both perceiving complex visual and language inputs, and understanding the underlying dynamics and causal relations. We also study an oracle model that explicitly combines these components via symbolic representations.

NeurIPS Conference 2020 Conference Paper

Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming

• Sumanth Dathathri
• Krishnamurthy Dvijotham
• Alexey Kurakin
• Aditi Raghunathan
• Jonathan Uesato
• Rudy R. Bunel
• Shreya Shankar
• Jacob Steinhardt

Convex relaxations have emerged as a promising approach for verifying properties of neural networks, but widely used using Linear Programming (LP) relaxations only provide meaningful certificates when networks are specifically trained to facilitate verification. This precludes many important applications which involve \emph{verification-agnostic} networks that are not trained specifically to promote verifiability. On the other hand, semidefinite programming (SDP) relaxations have shown success on verification-agnostic networks, such as adversarially trained image classifiers without additional regularization, but do not currently scale beyond small networks due to poor time and space asymptotics. In this work, we propose a first-order dual SDP algorithm that provides (1) any-time bounds (2) requires memory only linear in the total number of network activations and (3) has per-iteration complexity that scales linearly with the complexity of a forward/backward pass through the network. By exploiting iterative eigenvector methods, we express all solver operations in terms of forward and backward passes through the network, enabling efficient use of hardware optimized for deep learning. This allows us to dramatically improve the magnitude of $\ell_\infty$ perturbations for which we can verify robustness verification-agnostic networks ($1\% \to 88\%$ on MNIST, $6\%\to 40\%$ on CIFAR-10). We also demonstrate tight verification for a quadratic stability specification for the decoder of a variational autoencoder.

UAI Conference 2020 Conference Paper

Lagrangian Decomposition for Neural Network Verification

• Rudy Bunel
• Alessandro De Palma
• Alban Desmaison
• Krishnamurthy Dvijotham
• Pushmeet Kohli
• Philip H. S. Torr
• M. Pawan Kumar

A fundamental component of neural network verification is the computation of bounds on the values their outputs can take. Previous methods have either used off-the-shelf solvers, discarding the problem structure, or relaxed the problem even further, making the bounds unnecessarily loose. We propose a novel approach based on Lagrangian Decomposition. Our formulation admits an efficient supergradient ascent algorithm, as well as an improved proximal algorithm. Both the algorithms offer three advantages: (i) they yield bounds that are provably at least as tight as previous dual algorithms relying on Lagrangian relaxations; (ii) they are based on operations analogous to forward/backward pass of neural networks layers and are therefore easily parallelizable, amenable to GPU implementation and able to take advantage of the convolutional structure of problems; and (iii) they allow for anytime stopping while still providing valid bounds. Empirically, we show that we obtain bounds comparable with off-the-shelf solvers in a fraction of their running time, and obtain tighter bounds in the same time as previous dual algorithms. This results in an overall speed-up when employing the bounds for formal verification. Code for our algorithms is available at https: //github. com/oval-group/decomposition-plnn-bounds.

ICLR Conference 2020 Conference Paper

Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs

• Aditya Paliwal
• Felix Gimeno
• Vinod Nair
• Yujia Li 0001
• Miles Lubin
• Pushmeet Kohli
• Oriol Vinyals

We present a deep reinforcement learning approach to minimizing the execution cost of neural network computation graphs in an optimizing compiler. Unlike earlier learning-based works that require training the optimizer on the same graph to be optimized, we propose a learning approach that trains an optimizer offline and then generalizes to previously unseen graphs without further training. This allows our approach to produce high-quality execution decisions on real-world TensorFlow graphs in seconds instead of hours. We consider two optimization tasks for computation graphs: minimizing running time and peak memory usage. In comparison to an extensive set of baselines, our approach achieves significant improvements over classical and other learning-based methods on these two tasks.

NeurIPS Conference 2020 Conference Paper

The Autoencoding Variational Autoencoder

• Taylan Cemgil
• Sumedh Ghaisas
• Krishnamurthy Dvijotham
• Sven Gowal
• Pushmeet Kohli

Does a Variational AutoEncoder (VAE) consistently encode typical samples generated from its decoder? This paper shows that the perhaps surprising answer to this question is `No'; a (nominally trained) VAE does not necessarily amortize inference for typical samples that it is capable of generating. We study the implications of this behaviour on the learned representations and also the consequences of fixing it by introducing a notion of self consistency. Our approach hinges on an alternative construction of the variational approximation distribution to the true posterior of an extended VAE model with a Markov chain alternating between the encoder and the decoder. The method can be used to train a VAE model from scratch or given an already trained VAE, it can be run as a post processing step in an entirely self supervised way without access to the original training data. Our experimental analysis reveals that encoders trained with our self-consistency approach lead to representations that are robust (insensitive) to perturbations in the input introduced by adversarial attacks. We provide experimental results on the ColorMnist and CelebA benchmark datasets that quantify the properties of the learned representations and compare the approach with a baseline that is specifically trained for the desired property.

ICLR Conference 2020 Conference Paper

Toward Evaluating Robustness of Deep Reinforcement Learning with Continuous Control

• Tsui-Wei Weng
• Krishnamurthy Dvijotham
• Jonathan Uesato
• Kai Xiao
• Sven Gowal
• Robert Stanforth
• Pushmeet Kohli

Deep reinforcement learning has achieved great success in many previously difficult reinforcement learning tasks, yet recent studies show that deep RL agents are also unavoidably susceptible to adversarial perturbations, similar to deep neural networks in classification tasks. Prior works mostly focus on model-free adversarial attacks and agents with discrete actions. In this work, we study the problem of continuous control agents in deep RL with adversarial attacks and propose the first two-step algorithm based on learned model dynamics. Extensive experiments on various MuJoCo domains (Cartpole, Fish, Walker, Humanoid) demonstrate that our proposed framework is much more effective and efficient than model-free based attacks baselines in degrading agent performance as well as driving agents to unsafe states.

ICLR Conference 2020 Conference Paper

Towards Verified Robustness under Text Deletion Interventions

• Johannes Welbl
• Po-Sen Huang
• Robert Stanforth
• Sven Gowal
• Krishnamurthy Dvijotham
• Martin Szummer
• Pushmeet Kohli

Neural networks are widely used in Natural Language Processing, yet despite their empirical successes, their behaviour is brittle: they are both over-sensitive to small input changes, and under-sensitive to deletions of large fractions of input text. This paper aims to tackle under-sensitivity in the context of natural language inference by ensuring that models do not become more confident in their predictions as arbitrary subsets of words from the input text are deleted. We develop a novel technique for formal verification of this specification for models based on the popular decomposable attention mechanism by employing the efficient yet effective interval bound propagation (IBP) approach. Using this method we can efficiently prove, given a model, whether a particular sample is free from the under-sensitivity problem. We compare different training methods to address under-sensitivity, and compare metrics to measure it. In our experiments on the SNLI and MNLI datasets, we observe that IBP training leads to a significantly improved verified accuracy. On the SNLI test set, we can verify 18.4% of samples, a substantial improvement over only 2.8% using standard training.

NeurIPS Conference 2020 Conference Paper

Training Generative Adversarial Networks by Solving Ordinary Differential Equations

• Chongli Qin
• Yan Wu
• Jost Tobias Springenberg
• Andy Brock
• Jeff Donahue
• Timothy Lillicrap
• Pushmeet Kohli

The instability of Generative Adversarial Network (GAN) training has frequently been attributed to gradient descent. Consequently, recent methods have aimed to tailor the models and training procedures to stabilise the discrete updates. In contrast, we study the continuous-time dynamics induced by GAN training. Both theory and toy experiments suggest that these dynamics are in fact surprisingly stable. From this perspective, we hypothesise that instabilities in training GANs arise from the integration error in discretising the continuous dynamics. We experimentally verify that well-known ODE solvers (such as Runge-Kutta) can stabilise training - when combined with a regulariser that controls the integration error. Our approach represents a radical departure from previous methods which typically use adaptive optimisation and stabilisation techniques that constrain the functional space (e. g. Spectral Normalisation). Evaluation on CIFAR-10 and ImageNet shows that our method outperforms several strong baselines, demonstrating its efficacy.

IJCAI Conference 2019 Conference Paper

A Dual Approach to Verify and Train Deep Networks

• Sven Gowal
• Krishnamurthy Dvijotham
• Robert Stanforth
• Timothy Mann
• Pushmeet Kohli

This paper addressed the problem of formally verifying desirable properties of neural networks, i. e. , obtaining provable guarantees that neural networks satisfy specifications relating their inputs and outputs (e. g. , robustness to bounded norm adversarial perturbations). Most previous work on this topic was limited in its applicability by the size of the network, network architecture and the complexity of properties to be verified. In contrast, our framework applies to a general class of activation functions and specifications. We formulate verification as an optimization problem (seeking to find the largest violation of the specification) and solve a Lagrangian relaxation of the optimization problem to obtain an upper bound on the worst case violation of the specification being verified. Our approach is anytime, i. e. , it can be stopped at any time and a valid bound on the maximum violation can be obtained. Finally, we highlight how this approach can be used to train models that are amenable to verification.

NeurIPS Conference 2019 Conference Paper

Adversarial Robustness through Local Linearization

• Chongli Qin
• James Martens
• Sven Gowal
• Dilip Krishnan
• Krishnamurthy Dvijotham
• Alhussein Fawzi
• Soham De
• Robert Stanforth

Adversarial training is an effective methodology for training deep neural networks that are robust against adversarial, norm-bounded perturbations. However, the computational cost of adversarial training grows prohibitively as the size of the model and number of input dimensions increase. Further, training against less expensive and therefore weaker adversaries produces models that are robust against weak attacks but break down under attacks that are stronger. This is often attributed to the phenomenon of gradient obfuscation; such models have a highly non-linear loss surface in the vicinity of training examples, making it hard for gradient-based attacks to succeed even though adversarial examples still exist. In this work, we introduce a novel regularizer that encourages the loss to behave linearly in the vicinity of the training data, thereby penalizing gradient obfuscation while encouraging robustness. We show via extensive experiments on CIFAR-10 and ImageNet, that models trained with our regularizer avoid gradient obfuscation and can be trained significantly faster than adversarial training. Using this regularizer, we exceed current state of the art and achieve 47% adversarial accuracy for ImageNet with L-infinity norm adversarial perturbations of radius 4/255 under an untargeted, strong, white-box attack. Additionally, we match state of the art results for CIFAR-10 at 8/255.

NeurIPS Conference 2019 Conference Paper

Are Labels Required for Improving Adversarial Robustness?

• Jean-Baptiste Alayrac
• Jonathan Uesato
• Po-Sen Huang
• Alhussein Fawzi
• Robert Stanforth
• Pushmeet Kohli

Recent work has uncovered the interesting (and somewhat surprising) finding that training models to be invariant to adversarial perturbations requires substantially larger datasets than those required for standard classification. This result is a key hurdle in the deployment of robust machine learning models in many real world applications where labeled data is expensive. Our main insight is that unlabeled data can be a competitive alternative to labeled data for training adversarially robust models. Theoretically, we show that in a simple statistical setting, the sample complexity for learning an adversarially robust model from unlabeled data matches the fully supervised case up to constant factors. On standard datasets like CIFAR- 10, a simple Unsupervised Adversarial Training (UAT) approach using unlabeled data improves robust accuracy by 21. 7% over using 4K supervised examples alone, and captures over 95% of the improvement from the same number of labeled examples. Finally, we report an improvement of 4% over the previous state-of-the- art on CIFAR-10 against the strongest known attack by using additional unlabeled data from the uncurated 80 Million Tiny Images dataset. This demonstrates that our finding extends as well to the more realistic case where unlabeled data is also uncurated, therefore opening a new avenue for improving adversarial training.

ICML Conference 2019 Conference Paper

CompILE: Compositional Imitation Learning and Execution

• Thomas Kipf
• Yujia Li 0001
• Hanjun Dai
• Vinícius Flores Zambaldi
• Alvaro Sanchez-Gonzalez
• Edward Grefenstette
• Pushmeet Kohli
• Peter W. Battaglia

We introduce Compositional Imitation Learning and Execution (CompILE): a framework for learning reusable, variable-length segments of hierarchically-structured behavior from demonstration data. CompILE uses a novel unsupervised, fully-differentiable sequence segmentation module to learn latent encodings of sequential data that can be re-composed and executed to perform new tasks. Once trained, our model generalizes to sequences of longer length and from environment instances not seen during training. We evaluate CompILE in a challenging 2D multi-task environment and a continuous control task, and show that it can find correct task boundaries and event encodings in an unsupervised manner. Latent codes and associated behavior policies discovered by CompILE can be used by a hierarchical agent, where the high-level policy selects actions in the latent code space, and the low-level, task-specific policies are simply the learned decoders. We found that our CompILE-based agent could learn given only sparse rewards, where agents without task-specific policies struggle.

UAI Conference 2019 Conference Paper

Efficient Neural Network Verification with Exactness Characterization

• Krishnamurthy Dvijotham
• Robert Stanforth
• Sven Gowal
• Chongli Qin
• Soham De
• Pushmeet Kohli

Remarkable progress has been made on verification of neural networks, i. e. , showing that neural networks are provably consistent with specifications encoding properties like adversarial robustness. Recent methods developed for scalable neural network verification are based on computing an upper bound on the worst-case violation of the specification. Semidefinite programming (SDP) has been proposed as a means to obtain tight upper bounds. However, SDP solvers do not scale to large neural networks. We introduce a Lagrangian relaxation based on the SDP formulation and a novel algorithm to solve the relaxation that scales to networks that are two orders of magnitude larger than the off-the-shelf SDP solvers. Although verification of neural networks is known to be NP-hard in general, we develop the first known sufficient conditions under which a polynomial time verification algorithm (based on the above relaxation) is guaranteed to perform exact verification (i. e. , either verify a property or establish it is untrue). The algorithm can be implemented using primitives available readily in common deep learning frameworks. Experiments show that the algorithm is fast, and is able to compute tight upper bounds on the error rates under adversarial attacks of convolutional networks trained on MNIST and CIFAR-10.

ICML Conference 2019 Conference Paper

Graph Matching Networks for Learning the Similarity of Graph Structured Objects

• Yujia Li 0001
• Chenjie Gu
• Thomas Dullien
• Oriol Vinyals
• Pushmeet Kohli

This paper addresses the challenging problem of retrieval and matching of graph structured objects, and makes two key contributions. First, we demonstrate how Graph Neural Networks (GNN), which have emerged as an effective model for various supervised prediction problems defined on structured data, can be trained to produce embedding of graphs in vector spaces that enables efficient similarity reasoning. Second, we propose a novel Graph Matching Network model that, given a pair of graphs as input, computes a similarity score between them by jointly reasoning on the pair through a new cross-graph attention-based matching mechanism. We demonstrate the effectiveness of our models on different domains including the challenging problem of control-flow graph based function similarity search that plays an important role in the detection of vulnerabilities in software systems. The experimental analysis demonstrates that our models are not only able to exploit structure in the context of similarity learning but they can also outperform domain specific baseline systems that have been carefully hand-engineered for these problems.

NeurIPS Conference 2019 Conference Paper

Learning Transferable Graph Exploration

• Hanjun Dai
• Yujia Li
• Chenglong Wang
• Rishabh Singh
• Po-Sen Huang
• Pushmeet Kohli

This paper considers the problem of efficient exploration of unseen environments, a key challenge in AI. We propose a learning to explore' framework where we learn a policy from a distribution of environments. At test time, presented with an unseen environment from the same distribution, the policy aims to generalize the exploration strategy to visit the maximum number of unique states in a limited number of steps. We particularly focus on environments with graph-structured state-spaces that are encountered in many important real-world applications like software testing and map building. We formulate this task as a reinforcement learning problem where the exploration' agent is rewarded for transitioning to previously unseen environment states and employ a graph-structured memory to encode the agent's past trajectory. Experimental results demonstrate that our approach is extremely effective for exploration of spatial maps; and when applied on the challenging problems of coverage-guided software-testing of domain-specific programs and real-world mobile applications, it outperforms methods that have been hand-engineered by human experts.

ICML Conference 2019 Conference Paper

Structured agents for physical construction

• Victor Bapst
• Alvaro Sanchez-Gonzalez
• Carl Doersch
• Kimberly L. Stachenfeld
• Pushmeet Kohli
• Peter W. Battaglia
• Jessica B. Hamrick

Physical construction—the ability to compose objects, subject to physical dynamics, to serve some function—is fundamental to human intelligence. We introduce a suite of challenging physical construction tasks inspired by how children play with blocks, such as matching a target configuration, stacking blocks to connect objects together, and creating shelter-like structures over target objects. We examine how a range of deep reinforcement learning agents fare on these challenges, and introduce several new approaches which provide superior performance. Our results show that agents which use structured representations (e. g. , objects and scene graphs) and structured policies (e. g. , object-centric actions) outperform those which use less structured representations, and generalize better beyond their training when asked to reason about larger scenes. Model-based agents which use Monte-Carlo Tree Search also outperform strictly model-free agents in our most challenging construction problems. We conclude that approaches which combine structured representations and reasoning with powerful learning are a key path toward agents that possess rich intuitive physics, scene understanding, and planning.

AAMAS Conference 2019 Conference Paper

The Body is Not a Given: Joint Agent Policy Learning and Morphology Evolution

• Dylan Banarse
• Yoram Bachrach
• Siqi Liu
• Guy Lever
• Nicolas Heess
• Chrisantha Fernando
• Pushmeet Kohli
• Thore Graepel

Reinforcement learning (RL) has proven to be a powerful paradigm for deriving complex behaviors from simple reward signals in a wide range of environments. When applying RL to continuous control agents in simulated physics environments, the body is usually considered to be part of the environment. However, during evolution the physical body of biological organisms and their controlling brains are co-evolved, thus exploring a much larger space of actuator/controller configurations. Put differently, the intelligence does not reside only in the agent’s mind, but also in the design of their body. We propose a method for uncovering strong agents, consisting of a good combination of a body and policy, based on combining RL with an evolutionary procedure. Given the resulting agent, we also propose an approach for identifying the body changes that contributed the most to the agent performance. We use the Shapley value from cooperative game theory to find the fair contribution of individual components, taking into account synergies between components. We evaluate our methods in an environment similar to the the recently proposed Robo-Sumo task, where agents in a software physics simulator compete in tipping over their opponent or pushing them out of the arena. Our results show that the proposed methods are indeed capable of generating strong agents, significantly outperforming baselines that focus on optimizing the agent policy alone. A video is available at: https: //youtu. be/CHlecRim9PI

ICLR Conference 2019 Conference Paper

The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision

• Jiayuan Mao
• Chuang Gan 0001
• Pushmeet Kohli
• Joshua B. Tenenbaum
• Jiajun Wu 0001

We propose the Neuro-Symbolic Concept Learner (NS-CL), a model that learns visual concepts, words, and semantic parsing of sentences without explicit supervision on any of them; instead, our model learns by simply looking at images and reading paired questions and answers. Our model builds an object-based scene representation and translates sentences into executable, symbolic programs. To bridge the learning of two modules, we use a neuro-symbolic reasoning module that executes these programs on the latent scene representation. Analogical to human concept learning, the perception module learns visual concepts based on the language description of the object being referred to. Meanwhile, the learned visual concepts facilitate learning new words and parsing new sentences. We use curriculum learning to guide the searching over the large compositional space of images and language. Extensive experiments demonstrate the accuracy and efficiency of our model on learning visual concepts, word representations, and semantic parsing of sentences. Further, our method allows easy generalization to new object attributes, compositions, language concepts, scenes and questions, and even new program domains. It also empowers applications including visual question answering and bidirectional image-text retrieval.

UAI Conference 2018 Conference Paper

A Dual Approach to Scalable Verification of Deep Networks

• Krishnamurthy Dvijotham
• Robert Stanforth
• Sven Gowal
• Timothy A. Mann
• Pushmeet Kohli

This paper addresses the problem of formally verifying desirable properties of neural networks, i. e. , obtaining provable guarantees that neural networks satisfy specifications relating their inputs and outputs (robustness to bounded norm adversarial perturbations, for example). Most previous work on this topic was limited in its applicability by the size of the network, network architecture and the complexity of properties to be verified. In contrast, our framework applies to a general class of activation functions and specifications on neural network inputs and outputs. We formulate verification as an optimization problem (seeking to find the largest violation of the specification) and solve a Lagrangian relaxation of the optimization problem to obtain an upper bound on the worst case violation of the specification being verified. Our approach is anytime i. e. it can be stopped at any time and a valid bound on the maximum violation can be obtained. We develop specialized verification algorithms with provable tightness guarantees under special assumptions and demonstrate the practical significance of our general verification approach on a variety of verification tasks.

NeurIPS Conference 2018 Conference Paper

A Unified View of Piecewise Linear Neural Network Verification

• Rudy Bunel
• Ilker Turkaslan
• Philip Torr
• Pushmeet Kohli
• Pawan Mudigonda

The success of Deep Learning and its potential use in many safety-critical applications has motivated research on formal verification of Neural Network (NN) models. Despite the reputation of learned NN models to behave as black boxes and the theoretical hardness of proving their properties, researchers have been successful in verifying some classes of models by exploiting their piecewise linear structure and taking insights from formal methods such as Satisifiability Modulo Theory. These methods are however still far from scaling to realistic neural networks. To facilitate progress on this crucial area, we make two key contributions. First, we present a unified framework that encompasses previous methods. This analysis results in the identification of new methods that combine the strengths of multiple existing approaches, accomplishing a speedup of two orders of magnitude compared to the previous state of the art. Second, we propose a new data set of benchmarks which includes a collection of previously released testcases. We use the benchmark to provide the first experimental comparison of existing algorithms and identify the factors impacting the hardness of verification problems.

ICML Conference 2018 Conference Paper

Adversarial Risk and the Dangers of Evaluating Against Weak Attacks

• Jonathan Uesato
• Brendan O'Donoghue
• Pushmeet Kohli
• Aäron van den Oord

This paper investigates recently proposed approaches for defending against adversarial examples and evaluating adversarial robustness. We motivate adversarial risk as an objective for achieving models robust to worst-case inputs. We then frame commonly used attacks and evaluation metrics as defining a tractable surrogate objective to the true adversarial risk. This suggests that models may optimize this surrogate rather than the true adversarial risk. We formalize this notion as obscurity to an adversary, and develop tools and heuristics for identifying obscured models and designing transparent models. We demonstrate that this is a significant problem in practice by repurposing gradient-free optimization techniques into adversarial attacks, which we use to decrease the accuracy of several recently proposed defenses to near zero. Our hope is that our formulations and results will help researchers to develop more powerful defenses.

NeurIPS Conference 2018 Conference Paper

Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding

• Kexin Yi
• Jiajun Wu
• Chuang Gan
• Antonio Torralba
• Pushmeet Kohli
• Josh Tenenbaum

We marry two powerful ideas: deep representation learning for visual recognition and language understanding, and symbolic program execution for reasoning. Our neural-symbolic visual question answering (NS-VQA) system first recovers a structural scene representation from the image and a program trace from the question. It then executes the program on the scene representation to obtain an answer. Incorporating symbolic structure as prior knowledge offers three unique advantages. First, executing programs on a symbolic space is more robust to long program traces; our model can solve complex reasoning tasks better, achieving an accuracy of 99. 8% on the CLEVR dataset. Second, the model is more data- and memory-efficient: it performs well after learning on a small number of training data; it can also encode an image into a compact representation, requiring less storage than existing methods for offline question answering. Third, symbolic program execution offers full transparency to the reasoning process; we are thus able to interpret and diagnose each execution step.

ICML Conference 2018 Conference Paper

Programmatically Interpretable Reinforcement Learning

• Abhinav Verma 0001
• Vijayaraghavan Murali
• Rishabh Singh
• Pushmeet Kohli
• Swarat Chaudhuri

We present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies. Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents policies using a high-level, domain-specific programming language. Such programmatic policies have the benefits of being more easily interpreted than neural networks, and being amenable to verification by symbolic methods. We propose a new method, called Neurally Directed Program Search (NDPS), for solving the challenging nonsmooth optimization problem of finding a programmatic policy with maximal reward. NDPS works by first learning a neural policy network using DRL, and then performing a local search over programmatic policies that seeks to minimize a distance from this neural “oracle”. We evaluate NDPS on the task of learning to drive a simulated car in the TORCS car-racing environment. We demonstrate that NDPS is able to discover human-readable policies that pass some significant performance bars. We also show that PIRL policies can have smoother trajectories, and can be more easily transferred to environments not encountered during training, than corresponding policies discovered by DRL.

ICML Conference 2017 Conference Paper

Batched High-dimensional Bayesian Optimization via Structural Kernel Learning

• Zi Wang 0004
• Chengtao Li
• Stefanie Jegelka
• Pushmeet Kohli

Optimization of high-dimensional black-box functions is an extremely challenging problem. While Bayesian optimization has emerged as a popular approach for optimizing black-box functions, its applicability has been limited to low-dimensional problems due to its computational and statistical challenges arising from high-dimensional settings. In this paper, we propose to tackle these challenges by (1) assuming a latent additive structure in the function and inferring it properly for more efficient and effective BO, and (2) performing multiple evaluations in parallel to reduce the number of iterations required by the method. Our novel approach learns the latent structure with Gibbs sampling and constructs batched queries using determinantal point processes. Experimental validations on both synthetic and real-world functions demonstrate that the proposed method outperforms the existing state-of-the-art approaches.

ICML Conference 2017 Conference Paper

Learning Continuous Semantic Representations of Symbolic Expressions

• Miltiadis Allamanis
• Pankajan Chanthirasegaran
• Pushmeet Kohli
• Charles Sutton

Combining abstract, symbolic reasoning with continuous neural reasoning is a grand challenge of representation learning. As a step in this direction, we propose a new architecture, called neural equivalence network, for the problem of learning continuous semantic representations of algebraic and logical expressions. These networks are trained to represent semantic equivalence, even of expressions that are syntactically very different. The challenge is that semantic representations must be computed in a syntax-directed manner, because semantics is compositional, but at the same time, small changes in syntax can lead to very large changes in semantics, which can be difficult for continuous neural architectures. We perform an exhaustive evaluation on the task of checking equivalence on a highly diverse class of symbolic algebraic and boolean expression types, showing that our model significantly outperforms existing architectures.

NeurIPS Conference 2017 Conference Paper

Learning Disentangled Representations with Semi-Supervised Deep Generative Models

• Siddharth N
• Brooks Paige
• Jan-Willem van de Meent
• Alban Desmaison
• Noah Goodman
• Pushmeet Kohli
• Frank Wood
• Philip Torr

Variational autoencoders (VAEs) learn representations of data by jointly training a probabilistic encoder and decoder network. Typically these models encode all features of the data into a single variable. Here we are interested in learning disentangled representations that encode distinct aspects of the data into separate variables. We propose to learn such representations using model architectures that generalise from standard VAEs, employing a general graphical model structure in the encoder and decoder. This allows us to train partially-specified models that make relatively strong assumptions about a subset of interpretable variables and rely on the flexibility of neural networks to learn representations for the remaining variables. We further define a general objective for semi-supervised learning in this model class, which can be approximated using an importance sampling procedure. We evaluate our framework's ability to learn disentangled representations, both by qualitative exploration of its generative capacity, and quantitative evaluation of its discriminative ability on a variety of models and datasets.

NeurIPS Conference 2017 Conference Paper

Learning to See Physics via Visual De-animation

• Jiajun Wu
• Erika Lu
• Pushmeet Kohli
• Bill Freeman
• Josh Tenenbaum

We introduce a paradigm for understanding physical scenes without human annotations. At the core of our system is a physical world representation that is first recovered by a perception module and then utilized by physics and graphics engines. During training, the perception module and the generative models learn by visual de-animation --- interpreting and reconstructing the visual information stream. During testing, the system first recovers the physical world state, and then uses the generative models for reasoning and future prediction. Even more so than forward simulation, inverting a physics or graphics engine is a computationally hard problem; we overcome this challenge by using a convolutional inversion network. Our system quickly recognizes the physical world state from appearance and motion cues, and has the flexibility to incorporate both differentiable and non-differentiable physics and graphics engines. We evaluate our system on both synthetic and real datasets involving multiple physical scenes, and demonstrate that our system performs well on both physical state estimation and reasoning problems. We further show that the knowledge learned on the synthetic dataset generalizes to constrained real images.

NeurIPS Conference 2017 Conference Paper

Neural Program Meta-Induction

• Jacob Devlin
• Rudy Bunel
• Rishabh Singh
• Matthew Hausknecht
• Pushmeet Kohli

Most recently proposed methods for Neural Program induction work under the assumption of having a large set of input/output (I/O) examples for learning any given input-output mapping. This paper aims to address the problem of data and computation efficiency of program induction by leveraging information from related tasks. Specifically, we propose two novel approaches for cross-task knowledge transfer to improve program induction in limited-data scenarios. In our first proposal, portfolio adaptation, a set of induction models is pretrained on a set of related tasks, and the best model is adapted towards the new task using transfer learning. In our second approach, meta program induction, a $k$-shot learning approach is used to make a model generalize to new tasks without additional training. To test the efficacy of our methods, we constructed a new benchmark of programs written in the Karel programming language. Using an extensive experimental evaluation on the Karel benchmark, we demonstrate that our proposals dramatically outperform the baseline induction method that does not use knowledge transfer. We also analyze the relative performance of the two approaches and study conditions in which they perform best. In particular, meta induction outperforms all existing approaches under extreme data sparsity (when a very small number of examples are available), i. e. , fewer than ten. As the number of available I/O examples increase (i. e. a thousand or more), portfolio adapted program induction becomes the best approach. For intermediate data sizes, we demonstrate that the combined method of adapted meta program induction has the strongest performance.

ICML Conference 2017 Conference Paper

RobustFill: Neural Program Learning under Noisy I/O

• Jacob Devlin
• Jonathan Uesato
• Surya Bhupatiraju
• Rishabh Singh
• Abdel-rahman Mohamed
• Pushmeet Kohli

The problem of automatically generating a computer program from some specification has been studied since the early days of AI. Recently, two competing approaches for `automatic program learning’ have received significant attention: (1) `neural program synthesis’, where a neural network is conditioned on input/output (I/O) examples and learns to generate a program, and (2) `neural program induction’, where a neural network generates new outputs directly using a latent program representation. Here, for the first time, we directly compare both approaches on a large-scale, real-world learning task and we additionally contrast to rule-based program synthesis, which uses hand-crafted semantics to guide the program generation. Our neural models use a modified attention RNN to allow encoding of variable-sized sets of I/O pairs, which achieve 92\% accuracy on a real-world test set, compared to the 34\% accuracy of the previous best neural synthesis approach. The synthesis model also outperforms a comparable induction model on this task, but we more importantly demonstrate that the strength of each approach is highly dependent on the evaluation metric and end-user application. Finally, we show that we can train our neural models to remain very robust to the type of noise expected in real-world data (e. g. , typos), while a highly-engineered rule-based system fails entirely.

ICML Conference 2017 Conference Paper

Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning

• Jakob N. Foerster
• Nantas Nardelli
• Gregory Farquhar
• Triantafyllos Afouras
• Philip H. S. Torr
• Pushmeet Kohli
• Shimon Whiteson

Many real-world problems, such as network packet routing and urban traffic control, are naturally modeled as multi-agent reinforcement learning (RL) problems. However, existing multi-agent RL methods typically scale poorly in the problem size. Therefore, a key challenge is to translate the success of deep learning on single-agent RL to the multi-agent setting. A major stumbling block is that independent Q-learning, the most popular multi-agent RL method, introduces nonstationarity that makes it incompatible with the experience replay memory on which deep Q-learning relies. This paper proposes two methods that address this problem: 1) using a multi-agent variant of importance sampling to naturally decay obsolete data and 2) conditioning each agent’s value function on a fingerprint that disambiguates the age of the data sampled from the replay memory. Results on a challenging decentralised variant of StarCraft unit micromanagement confirm that these methods enable the successful combination of experience replay with multi-agent RL.

ICML Conference 2017 Conference Paper

Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning

• Junhyuk Oh
• Satinder Singh 0001
• Honglak Lee
• Pushmeet Kohli

As a step towards developing zero-shot task generalization capabilities in reinforcement learning (RL), we introduce a new RL problem where the agent should learn to execute sequences of instructions after learning useful skills that solve subtasks. In this problem, we consider two types of generalizations: to previously unseen instructions and to longer sequences of instructions. For generalization over unseen instructions, we propose a new objective which encourages learning correspondences between similar subtasks by making analogies. For generalization over sequential instructions, we present a hierarchical architecture where a meta controller learns to use the acquired skills for executing the instructions. To deal with delayed reward, we propose a new neural architecture in the meta controller that learns when to update the subtask, which makes learning more efficient. Experimental results on a stochastic 3D domain show that the proposed ideas are crucial for generalization to longer instructions as well as unseen instructions.

NeurIPS Conference 2016 Conference Paper

Adaptive Neural Compilation

• Rudy Bunel
• Alban Desmaison
• Pawan Mudigonda
• Pushmeet Kohli
• Philip Torr

This paper proposes an adaptive neural-compilation framework to address the problem of learning efficient program. Traditional code optimisation strategies used in compilers are based on applying pre-specified set of transformations that make the code faster to execute without changing its semantics. In contrast, our work involves adapting programs to make them more efficient while considering correctness only on a target input distribution. Our approach is inspired by the recent works on differentiable representations of programs. We show that it is possible to compile programs written in a low-level language to a differentiable representation. We also show how programs in this representation can be optimised to make them efficient on a target distribution of inputs. Experimental results demonstrate that our approach enables learning specifically-tuned algorithms for given data distributions with a high success rate.

NeurIPS Conference 2016 Conference Paper

Batched Gaussian Process Bandit Optimization via Determinantal Point Processes

• Tarun Kathuria
• Amit Deshpande
• Pushmeet Kohli

Gaussian Process bandit optimization has emerged as a powerful tool for optimizing noisy black box functions. One example in machine learning is hyper-parameter optimization where each evaluation of the target function may require training a model which may involve days or even weeks of computation. Most methods for this so-called “Bayesian optimization” only allow sequential exploration of the parameter space. However, it is often desirable to propose batches or sets of parameter values to explore simultaneously, especially when there are large parallel processing facilities at our disposal. Batch methods require modeling the interaction between the different evaluations in the batch, which can be expensive in complex scenarios. In this paper, we propose a new approach for parallelizing Bayesian optimization by modeling the diversity of a batch via Determinantal point processes (DPPs) whose kernels are learned automatically. This allows us to generalize a previous result as well as prove better regret bounds based on DPP sampling. Our experiments on a variety of synthetic and real-world robotics and hyper-parameter optimization tasks indicate that our DPP-based methods, especially those based on DPP sampling, outperform state-of-the-art methods.

NeurIPS Conference 2016 Conference Paper

PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

• Mikhail Figurnov
• Aizhan Ibraimova
• Dmitry Vetrov
• Pushmeet Kohli

We propose a novel approach to reduce the computational cost of evaluation of convolutional neural networks, a factor that has hindered their deployment in low-power devices such as mobile phones. Inspired by the loop perforation technique from source code optimization, we speed up the bottleneck convolutional layers by skipping their evaluation in some of the spatial positions. We propose and analyze several strategies of choosing these positions. We demonstrate that perforation can accelerate modern convolutional networks such as AlexNet and VGG-16 by a factor of 2x - 4x. Additionally, we show that perforation is complementary to the recently proposed acceleration method of Zhang et al.

UAI Conference 2016 Conference Paper

Political Dimensionality Estimation Using a Probabilistic Graphical Model

• Yoad Lewenberg
• Yoram Bachrach
• Lucas Bordeaux
• Pushmeet Kohli

This paper attempts to move beyond the left-right characterization of political ideologies. We propose a trait based probabilistic model for estimating the manifold of political opinion. We demonstrate the efficacy of our model on two novel and large scale datasets of public opinion. Our experiments show that although the political spectrum is richer than a simple left-right structure, peoples’ opinions on seemingly unrelated political issues are very correlated, so fewer than 10 dimensions are enough to represent peoples’ entire political opinion.

JAIR Journal 2016 Journal Article

Time-Sensitive Bayesian Information Aggregation for Crowdsourcing Systems

• Matteo Venanzi
• John Guiver
• Pushmeet Kohli
• Nicholas R. Jennings

Many aspects of the design of efficient crowdsourcing processes, such as defining worker’s bonuses, fair prices and time limits of the tasks, involve knowledge of the likely duration of the task at hand. In this work we introduce a new time–sensitive Bayesian aggregation method that simultaneously estimates a task’s duration and obtains reliable aggregations of crowdsourced judgments. Our method, called BCCTime, uses latent variables to represent the uncertainty about the workers’ completion time, the tasks’ duration and the workers’ accuracy. To relate the quality of a judgment to the time a worker spends on a task, our model assumes that each task is completed within a latent time window within which all workers with a propensity to genuinely attempt the labelling task (i.e., no spammers) are expected to submit their judgments. In contrast, workers with a lower propensity to valid labelling, such as spammers, bots or lazy labellers, are assumed to perform tasks considerably faster or slower than the time required by normal workers. Specifically, we use efficient message-passing Bayesian inference to learn approximate posterior probabilities of (i) the confusion matrix of each worker, (ii) the propensity to valid labelling of each worker, (iii) the unbiased duration of each task and (iv) the true label of each task. Using two real- world public datasets for entity linking tasks, we show that BCCTime produces up to 11% more accurate classifications and up to 100% more informative estimates of a task’s duration compared to state–of–the–art methods.

NeurIPS Conference 2015 Conference Paper

Deep Convolutional Inverse Graphics Network

• Tejas Kulkarni
• William Whitney
• Pushmeet Kohli
• Josh Tenenbaum

This paper presents the Deep Convolution Inverse Graphics Network (DC-IGN), a model that aims to learn an interpretable representation of images, disentangled with respect to three-dimensional scene structure and viewing transformations such as depth rotations and lighting variations. The DC-IGN model is composed of multiple layers of convolution and de-convolution operators and is trained using the Stochastic Gradient Variational Bayes (SGVB) algorithm. We propose a training procedure to encourage neurons in the graphics code layer to represent a specific transformation (e. g. pose or light). Given a single input image, our model can generate new images of the same object with variations in pose and lighting. We present qualitative and quantitative tests of the model's efficacy at learning a 3D rendering engine for varied object classes including faces and chairs.

NeurIPS Conference 2015 Conference Paper

Efficient Non-greedy Optimization of Decision Trees

• Mohammad Norouzi
• Maxwell Collins
• Matthew Johnson
• David Fleet
• Pushmeet Kohli

Decision trees and randomized forests are widely used in computer vision and machine learning. Standard algorithms for decision tree induction optimize the split functions one node at a time according to some splitting criteria. This greedy procedure often leads to suboptimal trees. In this paper, we present an algorithm for optimizing the split functions at all levels of the tree jointly with the leaf parameters, based on a global objective. We show that the problem of finding optimal linear-combination (oblique) splits for decision trees is related to structured prediction with latent variables, and we formulate a convex-concave upper bound on the tree's empirical loss. Computing the gradient of the proposed surrogate objective with respect to each training exemplar is O(d^2), where d is the tree depth, and thus training deep trees is feasible. The use of stochastic gradient descent for optimization enables effective training with large datasets. Experiments on several classification benchmarks demonstrate that the resulting non-greedy decision trees outperform greedy decision tree baselines.

IJCAI Conference 2015 Conference Paper

Information Gathering in Networks via Active Exploration

• Adish Singla
• Eric Horvitz
• Pushmeet Kohli
• Ryen White
• Andreas Krause

How should we gather information in a network, where each node’s visibility is limited to its local neighborhood? This problem arises in numerous real-world applications, such as surveying and task routing in social networks, team formation in collaborative networks and experimental design with dependency constraints. Often the informativeness of a set of nodes can be quantified via a submodular utility function. Existing approaches for submodular optimization, however, require that the set of all nodes that can be selected is known ahead of time, which is often unrealistic. In contrast, we propose a novel model where we start our exploration from an initial node, and new nodes become visible and available for selection only once one of their neighbors has been chosen. We then present a general algorithm NETEXP for this problem, and provide theoretical bounds on its performance dependent on structural properties of the underlying network. We evaluate our methodology on various simulated problem instances as well as on data collected from social question answering system deployed within a large enterprise.

UAI Conference 2015 Conference Paper

Minimizing Expected Losses in Perturbation Models with Multidimensional Parametric Min-cuts

• Adrian Kim
• Kyomin Jung
• Yongsub Lim
• Daniel Tarlow
• Pushmeet Kohli

We consider the problem of learning perturbation-based probabilistic models by computing and differentiating expected losses. This is a challenging computational problem that has traditionally been tackled using Monte Carlo-based methods. In this work, we show how a generalization of parametric min-cuts can be used to address the same problem, achieving higher accuracy and faster performance than a sampling-based baseline. Utilizing our proposed Skeleton Method, we show that we can learn the perturbation model so as to directly minimize expected losses. Experimental results show that this approach offers promise as a new way of training structured prediction models under complex loss functions.

NeurIPS Conference 2014 Conference Paper

Just-In-Time Learning for Fast and Flexible Inference

• S. M. Ali Eslami
• Daniel Tarlow
• Pushmeet Kohli
• John Winn

Much of research in machine learning has centered around the search for inference algorithms that are both general-purpose and efficient. The problem is extremely challenging and general inference remains computationally expensive. We seek to address this problem by observing that in most specific applications of a model, we typically only need to perform a small subset of all possible inference computations. Motivated by this, we introduce just-in-time learning, a framework for fast and flexible inference that learns to speed up inference at run-time. Through a series of experiments, we show how this framework can allow us to combine the flexibility of sampling with the efficiency of deterministic message-passing.

AAAI Conference 2013 Conference Paper

A Fast Bandit Algorithm for Recommendation to Users With Heterogenous Tastes

• Pushmeet Kohli
• Mahyar Salek
• Greg Stoddard

We study recommendation in scenarios where there’s no prior information about the quality of content in the system. We present an online algorithm that continually optimizes recommendation relevance based on behavior of past users. Our method trades weaker theoretical guarantees in asymptotic performance than the state-ofthe-art for stronger theoretical guarantees in the online setting. We test our algorithm on real-world data collected from previous recommender systems and show that our algorithm learns faster than existing methods and performs equally well in the long-run.

NeurIPS Conference 2013 Conference Paper

Decision Jungles: Compact and Rich Models for Classification

• Jamie Shotton
• Toby Sharp
• Pushmeet Kohli
• Sebastian Nowozin
• John Winn
• Antonio Criminisi

Randomized decision trees and forests have a rich history in machine learning and have seen considerable success in application, perhaps particularly so for computer vision. However, they face a fundamental limitation: given enough data, the number of nodes in decision trees will grow exponentially with depth. For certain applications, for example on mobile or embedded processors, memory is a limited resource, and so the exponential growth of trees limits their depth, and thus their potential accuracy. This paper proposes decision jungles, revisiting the idea of ensembles of rooted decision directed acyclic graphs (DAGs), and shows these to be compact and powerful discriminative models for classification. Unlike conventional decision trees that only allow one path to every node, a DAG in a decision jungle allows multiple paths from the root to each leaf. We present and compare two new node merging algorithms that jointly optimize both the features and the structure of the DAGs efficiently. During training, node splitting and node merging are driven by the minimization of exactly the same objective function, here the weighted sum of entropies at the leaves. Results on varied datasets show that, compared to decision forests and several other baselines, decision jungles require dramatically less memory while considerably improving generalization.

AAAI Conference 2013 Conference Paper

Optimal Coalition Structure Generation in Cooperative Graph Games

• Yoram Bachrach
• Pushmeet Kohli
• Vladimir Kolmogorov
• Morteza Zadimoghaddam

Representation languages for coalitional games are a key research area in algorithmic game theory. There is an inherent tradeoff between how general a language is, allowing it to capture more elaborate games, and how hard it is computationally to optimize and solve such games. One prominent such language is the simple yet expressive Weighted Graph Games (WGGs) representation (Deng and Papadimitriou 1994), which maintains knowledge about synergies between agents in the form of an edge weighted graph. We consider the problem of finding the optimal coalition structure in WGGs. The agents in such games are vertices in a graph, and the value of a coalition is the sum of the weights of the edges present between coalition members. The optimal coalition structure is a partition of the agents to coalitions, that maximizes the sum of utilities obtained by the coalitions. We show that finding the optimal coalition structure is not only hard for general graphs, but is also intractable for restricted families such as planar graphs which are amenable for many other combinatorial problems. We then provide algorithms with constant factor approximations for planar, minorfree and bounded degree graphs.

NeurIPS Conference 2012 Conference Paper

Context-Sensitive Decision Forests for Object Detection

• Peter Kontschieder
• Samuel Bulò
• Antonio Criminisi
• Pushmeet Kohli
• Marcello Pelillo
• Horst Bischof

In this paper we introduce Context-Sensitive Decision Forests - A new perspective to exploit contextual information in the popular decision forest framework for the object detection problem. They are tree-structured classifiers with the ability to access intermediate prediction (here: classification and regression) information during training and inference time. This intermediate prediction is available to each sample, which allows us to develop context-based decision criteria, used for refining the prediction process. In addition, we introduce a novel split criterion which in combination with a priority based way of constructing the trees, allows more accurate regression mode selection and hence improves the current context information. In our experiments, we demonstrate improved results for the task of pedestrian detection on the challenging TUD data set when compared to state-of-the-art methods.

NeurIPS Conference 2012 Conference Paper

Multiple Choice Learning: Learning to Produce Multiple Structured Outputs

• Abner Guzmán-rivera
• Dhruv Batra
• Pushmeet Kohli

The paper addresses the problem of generating multiple hypotheses for prediction tasks that involve interaction with users or successive components in a cascade. Given a set of multiple hypotheses, such components/users have the ability to automatically rank the results and thus retrieve the best one. The standard approach for handling this scenario is to learn a single model and then produce M-best Maximum a Posteriori (MAP) hypotheses from this model. In contrast, we formulate this multiple {\em choice} learning task as a multiple-output structured-output prediction problem with a loss function that captures the natural setup of the problem. We present a max-margin formulation that minimizes an upper-bound on this loss-function. Experimental results on the problems of image co-segmentation and protein side-chain prediction show that our method outperforms conventional approaches used for this scenario and leads to substantial improvements in prediction accuracy.

ICML Conference 2011 Conference Paper

Dynamic Tree Block Coordinate Ascent

• Daniel Tarlow
• Dhruv Batra
• Pushmeet Kohli
• Vladimir Kolmogorov

NeurIPS Conference 2011 Conference Paper

Higher-Order Correlation Clustering for Image Segmentation

• Sungwoong Kim
• Sebastian Nowozin
• Pushmeet Kohli
• Chang Yoo

For many of the state-of-the-art computer vision algorithms, image segmentation is an important preprocessing step. As such, several image segmentation algorithms have been proposed, however, with certain reservation due to high computational load and many hand-tuning parameters. Correlation clustering, a graph-partitioning algorithm often used in natural language processing and document clustering, has the potential to perform better than previously proposed image segmentation algorithms. We improve the basic correlation clustering formulation by taking into account higher-order cluster relationships. This improves clustering in the presence of local boundary ambiguities. We first apply the pairwise correlation clustering to image segmentation over a pairwise superpixel graph and then develop higher-order correlation clustering over a hypergraph that considers higher-order relations among superpixels. Fast inference is possible by linear programming relaxation, and also effective parameter learning framework by structured support vector machine is possible. Experimental results on various datasets show that the proposed higher-order correlation clustering outperforms other state-of-the-art image segmentation algorithms.

AAMAS Conference 2011 Conference Paper

Rip-off: Playing the Cooperative Negotiation Game

• Yoram Bachrach
• Pushmeet Kohli
• Thore Graepel

We propose "Rip-off", a new multi-player bargaining game based on the well-studied weighted voting game (WVG) model from cooperative game theory. Many different solution concepts, such as the Core and the Shapley value have been proposed to analyze models such as WVGs. However, there is little work on analyzing how humans actually play in such settings. We conducted several experiments where we let humans play "Rip-off". Our analysis reveals that although solutions of games played by humans do suffer from certain biases, a player's average payoff over several games is roughly reflected by the Shapley value.

AAMAS Conference 2011 Conference Paper

Team Coverage Games

• Yoram Bachrach
• Pushmeet Kohli
• Vladimir Kolmogorov

Team Coverage Games (TCGs) are a representation of cooperative games, where the value a coalition generates depends on both individual contributions of its members and synergies between them. The synergies are expressed in terms of the importance of the agents in various teams. TCGs model the synergy as a reduction in utility that occurs when team members are missing, causing the team not to achieve its full potential. We focus on the case where the utility reduction incured is a concave function of the importance of the missing team members and analyze the domain from a computational game theoretic perspective.

AAAI Conference 2010 Conference Paper

Coalitional Structure Generation in Skill Games

• Yoram Bachrach
• Reshef Meir
• Kyomin Jung
• Pushmeet Kohli

We consider optimizing the coalition structure in Coalitional Skill Games (CSGs), a succinct representation of coalitional games (Bachrach and Rosenschein 2008). In CSGs, the value of a coalition depends on the tasks its members can achieve. The tasks require various skills to complete them, and agents may have different skill sets. The optimal coalition structure is a partition of the agents to coalitions, that maximizes the sum of utilities obtained by the coalitions. We show that CSGs can represent any characteristic function, and consider optimal coalition structure generation in this representation. We provide hardness results, showing that in general CSGs, as well as in very restricted versions of them, computing the optimal coalition structure is hard. On the positive side, we show that the problem can be reformulated as constraint satisfaction on a hyper graph, and present an algorithm that finds the optimal coalition structure in polynomial time for instances with bounded tree-width and number of tasks.

UAI Conference 2010 Conference Paper

Exact and Approximate Inference in Associative Hierarchical Networks using Graph Cuts

• Chris Russell 0001
• Lubor Ladicky
• Pushmeet Kohli
• Philip H. S. Torr

Markov Networks are widely used through out computer vision and machine learning. An important subclass are the Associative Markov Networks which are used in a wide variety of applications. For these networks a good approximate minimum cost solution can be found efficiently using graph cut based move making algorithms such as alphaexpansion. Recently a related model has been proposed, the associative hierarchical network, which provides a natural generalisation of the Associative Markov Network for higher order cliques (i.e. clique size greater than two). This method provides a good model for object class segmentation problem in computer vision. Within this paper we briefly describe the associative hierarchical network and provide a computationally efficient method for approximate inference based on graph cuts. Our method performs well for networks containing hundreds of thousand of variables, and higher order potentials are defined over cliques containing tens of thousands of variables. Due to the size of these problems standard linear programming techniques are inapplicable. We show that our method has a bound of 4 for the solution of general associative hierarchical network with arbitrary clique size noting that few results on bounds exist for the solution of labelling of Markov Networks with higher order cliques.

NeurIPS Conference 2009 Conference Paper

Local Rules for Global MAP: When Do They Work ?

• Kyomin Jung
• Pushmeet Kohli
• Devavrat Shah

We consider the question of computing Maximum A Posteriori (MAP) assignment in an arbitrary pair-wise Markov Random Field (MRF). We present a randomized iterative algorithm based on simple local updates. The algorithm, starting with an arbitrary initial assignment, updates it in each iteration by first, picking a random node, then selecting an (appropriately chosen) random local neighborhood and optimizing over this local neighborhood. Somewhat surprisingly, we show that this algorithm finds a near optimal assignment within $2n\ln n$ iterations on average and with high probability for {\em any} $n$ node pair-wise MRF with {\em geometry} (i. e. MRF graph with polynomial growth) with the approximation error depending on (in a reasonable manner) the geometric growth rate of the graph and the average radius of the local neighborhood -- this allows for a graceful tradeoff between the complexity of the algorithm and the approximation error. Through extensive simulations, we show that our algorithm finds extremely good approximate solutions for various kinds of MRFs with geometry.

ICML Conference 2008 Conference Paper

On partial optimality in multi-label MRFs

• Pushmeet Kohli
• Alexander Shekhovtsov 0001
• Carsten Rother
• Vladimir Kolmogorov
• Philip H. S. Torr