Arrow Research search

Author name cluster

Carlo Ciliberto

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

36 papers
2 author rows

Possible papers

36

TMLR Journal 2024 Journal Article

Deep Tabular Learning via Distillation and Language Guidance

  • Ruohan Wang
  • Wenhao Fu
  • Carlo Ciliberto

Tabular data is arguably one of the most ubiquitous data structures in application domains such as science, healthcare, finance and manufacturing. Given the recent success of deep learning (DL), there has been a surge of new DL models for tabular learning. However, despite the efforts, tabular DL models still clearly trail behind tree-based approaches. In this work, we propose DisTab, a novel framework for tabular learning based on the transformer architecture. Our method leverages model distillation to mimic the favorable inductive biases of tree-based models, and incorporates language guidance for more expressive feature embeddings. Empirically, DisTab outperforms existing tabular DL models and is highly competitive against tree-based models across diverse datasets, effectively closing the gap with these methods.

PDF Details

NeurIPS Conference 2024 Conference Paper

Operator World Models for Reinforcement Learning

  • Pietro Novelli
  • Marco Pratticò
  • Massimiliano Pontil
  • Carlo Ciliberto

Policy Mirror Descent (PMD) is a powerful and theoretically sound methodology for sequential decision-making. However, it is not directly applicable to Reinforcement Learning (RL) due to the inaccessibility of explicit action-value functions. We address this challenge by introducing a novel approach based on learning a world model of the environment using conditional mean embeddings. Leveraging tools from operator theory we derive a closed-form expression of the action-value function in terms of the world model via simple matrix operations. Combining these estimators with PMD leads to POWR, a new RL algorithm for which we prove convergence rates to the global optimum. Preliminary experiments in finite and infinite state settings support the effectiveness of our method.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Conditional Meta-Learning of Linear Representations

  • Giulia Denevi
  • Massimiliano Pontil
  • Carlo Ciliberto

Standard meta-learning for representation learning aims to find a common representation to be shared across multiple tasks. The effectiveness of these methods is often limited when the nuances of the tasks’ distribution cannot be captured by a single representation. In this work we overcome this issue by inferring a conditioning function, mapping the tasks’ side information (such as the tasks’ training dataset itself) into a representation tailored to the task at hand. We study environments in which our conditional strategy outperforms standard meta-learning, such as those in which tasks can be organized in separate clusters according to the representation they share. We then propose a meta-algorithm capable of leveraging this advantage in practice. In the unconditional setting, our method yields a new estimator enjoying faster learning rates and requiring less hyper-parameters to tune than current state-of-the-art methods. Our results are supported by preliminary experiments.

PDF Details

ICML Conference 2022 Conference Paper

Distribution Regression with Sliced Wasserstein Kernels

  • Dimitri Meunier
  • Massimiliano Pontil
  • Carlo Ciliberto

The problem of learning functions over spaces of probabilities - or distribution regression - is gaining significant interest in the machine learning community. The main challenge in these settings is to identify a suitable representation capturing all relevant properties of a distribution. The well-established approach in this sense is to use kernel mean embeddings, which lift kernel-induced similarity on the input domain at the probability level. This strategy effectively tackles the two-stage sampling nature of the problem, enabling one to derive estimators with strong statistical guarantees, such as universal consistency and excess risk bounds. However, kernel mean embeddings implicitly hinge on the maximum mean discrepancy (MMD), a metric on probabilities, which is not the most suited to capture geometrical relations between distributions. In contrast, optimal transport (OT) metrics, are potentially more appealing. In this work, we propose an OT-based estimator for distribution regression. We build on the Sliced Wasserstein distance to obtain an OT-based representation. We study the theoretical properties of a kernel ridge regression estimator based on such representation, for which we prove universal consistency and excess risk bounds. Preliminary experiments complement our theoretical findings by showing the effectiveness of the proposed approach and compare it with MMD-based estimators.

Details

UAI Conference 2022 Conference Paper

Implicit kernel meta-learning using kernel integral forms

  • John Isak Texas Falk
  • Carlo Ciliberto
  • Massimiliano Pontil

Meta-learning algorithms have made significant progress in the context of meta-learning for image classification but less attention has been given to the regression setting. In this paper we propose to learn the probability distribution representing a random feature kernel that we wish to use within kernel ridge regression (KRR). We introduce two instances of this meta-learning framework, learning a neural network pushforward for a translation-invariant kernel and an affine pushforward for a neural network random feature kernel, both mapping from a Gaussian latent distribution. We learn the parameters of the pushforward by minimizing a meta-loss associated to the KRR objective. Since the resulting kernel does not admit an analytical form, we adopt a random feature sampling approach to approximate it. We call the resulting method Implicit Kernel Meta-Learning (IKML). We derive a meta-learning bound for IKML, which shows the role played by the number of tasks $T$, the task sample size $n$, and the number of random features $M$. In particular the bound implies that $M$ can be the chosen independently of $T$ and only mildly dependent on $n$. We introduce one synthetic and two real-world meta-learning regression benchmark datasets. Experiments on these datasets show that IKML

Details

NeurIPS Conference 2022 Conference Paper

Learning Dynamical Systems via Koopman Operator Regression in Reproducing Kernel Hilbert Spaces

  • Vladimir Kostic
  • Pietro Novelli
  • Andreas Maurer
  • Carlo Ciliberto
  • Lorenzo Rosasco
  • Massimiliano Pontil

We study a class of dynamical systems modelled as stationary Markov chains that admit an invariant distribution via the corresponding transfer or Koopman operator. While data-driven algorithms to reconstruct such operators are well known, their relationship with statistical learning is largely unexplored. We formalize a framework to learn the Koopman operator from finite data trajectories of the dynamical system. We consider the restriction of this operator to a reproducing kernel Hilbert space and introduce a notion of risk, from which different estimators naturally arise. We link the risk with the estimation of the spectral decomposition of the Koopman operator. These observations motivate a reduced-rank operator regression (RRR) estimator. We derive learning bounds for the proposed estimator, holding both in i. i. d and non i. i. d. settings, the latter in terms of mixing coefficients. Our results suggest RRR might be beneficial over other widely used estimators as confirmed in numerical experiments both for forecasting and mode decomposition.

PDF Details

ICML Conference 2022 Conference Paper

Measuring dissimilarity with diffeomorphism invariance

  • Théophile Cantelobre
  • Carlo Ciliberto
  • Benjamin Guedj
  • Alessandro Rudi

Measures of similarity (or dissimilarity) are a key ingredient to many machine learning algorithms. We introduce DID, a pairwise dissimilarity measure applicable to a wide range of data spaces, which leverages the data’s internal structure to be invariant to diffeomorphisms. We prove that DID enjoys properties which make it relevant for theoretical study and practical use. By representing each datum as a function, DID is defined as the solution to an optimization problem in a Reproducing Kernel Hilbert Space and can be expressed in closed-form. In practice, it can be efficiently approximated via Nystr{ö}m sampling. Empirical experiments support the merits of DID.

Details

ICRA Conference 2022 Conference Paper

Modular Adaptive Policy Selection for Multi- Task Imitation Learning through Task Division

  • Dafni Antotsiou
  • Carlo Ciliberto
  • Tae-Kyun Kim 0001

Deep imitation learning requires many expert demonstrations, which can be hard to obtain, especially when many tasks are involved. However, different tasks often share similarities, so learning them jointly can greatly benefit them and alleviate the need for many demonstrations. But, joint multi-task learning often suffers from negative transfer, sharing information that should be task-specific. In this work, we introduce a method to perform multi-task imitation while allowing for task-specific features. This is done by using proto-policies as modules to divide the tasks into simple sub-behaviours that can be shared. The proto-policies operate in parallel and are adaptively chosen by a selector mechanism that is jointly trained with the modules. Experiments on different sets of tasks show that our method improves upon the accuracy of single agents, task-conditioned and multi-headed multi-task agents, as well as state-of-the-art meta learning agents. We also demonstrate its ability to autonomously divide the tasks into both shared and task-specific sub-behaviours.

Details

ICRA Conference 2021 Conference Paper

Adversarial Imitation Learning with Trajectorial Augmentation and Correction

  • Dafni Antotsiou
  • Carlo Ciliberto
  • Tae-Kyun Kim 0001

Deep Imitation Learning requires a large number of expert demonstrations, which are not always easy to obtain, especially for complex tasks. A way to overcome this shortage of labels is through data augmentation. However, this cannot be easily applied to control tasks due to the sequential nature of the problem. In this work, we introduce a novel augmentation method which preserves the success of the augmented trajectories. To achieve this, we introduce a semi-supervised correction network that aims to correct distorted expert actions. To adequately test the abilities of the correction network, we develop an adversarial data augmented imitation architecture to train an imitation agent using synthetic experts. Additionally, we introduce a metric to measure diversity in trajectory datasets. Experiments show that our data augmentation strategy can improve accuracy and convergence time of adversarial imitation while preserving the diversity between the generated and real trajectories.

Details

NeurIPS Conference 2021 Conference Paper

PSD Representations for Effective Probability Models

  • Alessandro Rudi
  • Carlo Ciliberto

Finding a good way to model probability densities is key to probabilistic inference. An ideal model should be able to concisely approximate any probability while being also compatible with two main operations: multiplications of two models (product rule) and marginalization with respect to a subset of the random variables (sum rule). In this work, we show that a recently proposed class of positive semi-definite (PSD) models for non-negative functions is particularly suited to this end. In particular, we characterize both approximation and generalization capabilities of PSD models, showing that they enjoy strong theoretical guarantees. Moreover, we show that we can perform efficiently both sum and product rule in closed form via matrix operations, enjoying the same versatility of mixture models. Our results open the way to applications of PSD models to density estimation, decision theory, and inference.

PDF Details

NeurIPS Conference 2021 Conference Paper

The Role of Global Labels in Few-Shot Classification and How to Infer Them

  • Ruohan Wang
  • Massimiliano Pontil
  • Carlo Ciliberto

Few-shot learning is a central problem in meta-learning, where learners must quickly adapt to new tasks given limited training data. Recently, feature pre-training has become a ubiquitous component in state-of-the-art meta-learning methods and is shown to provide significant performance improvement. However, there is limited theoretical understanding of the connection between pre-training and meta-learning. Further, pre-training requires global labels shared across tasks, which may be unavailable in practice. In this paper, we show why exploiting pre-training is theoretically advantageous for meta-learning, and in particular the critical role of global labels. This motivates us to propose Meta Label Learning (MeLa), a novel meta-learning framework that automatically infers global labels to obtains robust few-shot models. Empirically, we demonstrate that MeLa is competitive with existing methods and provide extensive ablation experiments to highlight its key properties.

PDF Details

JMLR Journal 2020 Journal Article

A General Framework for Consistent Structured Prediction with Implicit Loss Embeddings

  • Carlo Ciliberto
  • Lorenzo Rosasco
  • Alessandro Rudi

We propose and analyze a novel theoretical and algorithmic framework for structured prediction. While so far the term has referred to discrete output spaces, here we consider more general settings, such as manifolds or spaces of probability measures. We define structured prediction as a problem where the output space lacks a vectorial structure. We identify and study a large class of loss functions that implicitly defines a suitable geometry on the problem. The latter is the key to develop an algorithmic framework amenable to a sharp statistical analysis and yielding efficient computations. When dealing with output spaces with infinite cardinality, a suitable implicit formulation of the estimator is shown to be crucial. [abs] [ pdf ][ bib ] &copy JMLR 2020. ( edit, beta )

PDF Details

NeurIPS Conference 2020 Conference Paper

Exploiting MMD and Sinkhorn Divergences for Fair and Transferable Representation Learning

  • Luca Oneto
  • Michele Donini
  • Giulia Luise
  • Carlo Ciliberto
  • Andreas Maurer
  • Massimiliano Pontil

Developing learning methods which do not discriminate subgroups in the population is a central goal of algorithmic fairness. One way to reach this goal is by modifying the data representation in order to meet certain fairness constraints. In this work we measure fairness according to demographic parity. This requires the probability of the possible model decisions to be independent of the sensitive information. We argue that the goal of imposing demographic parity can be substantially facilitated within a multitask learning setting. We present a method for learning a shared fair representation across multiple tasks, by means of different new constraints based on MMD and Sinkhorn Divergences. We derive learning bounds establishing that the learned representation transfers well to novel tasks. We present experiments on three real world datasets, showing that the proposed method outperforms state-of-the-art approaches by a significant margin.

PDF Details

NeurIPS Conference 2020 Conference Paper

Structured Prediction for Conditional Meta-Learning

  • Ruohan Wang
  • Yiannis Demiris
  • Carlo Ciliberto

The goal of optimization-based meta-learning is to find a single initialization shared across a distribution of tasks to speed up the process of learning new tasks. Conditional meta-learning seeks task-specific initialization to better capture complex task distributions and improve performance. However, many existing conditional methods are difficult to generalize and lack theoretical guarantees. In this work, we propose a new perspective on conditional meta-learning via structured prediction. We derive task-adaptive structured meta-learning (TASML), a principled framework that yields task-specific objective functions by weighing meta-training data on target tasks. Our non-parametric approach is model-agnostic and can be combined with existing meta-learning methods to achieve conditioning. Empirically, we show that TASML improves the performance of existing meta-learning models, and outperforms the state-of-the-art on benchmark datasets.

PDF Details

NeurIPS Conference 2020 Conference Paper

The Advantage of Conditional Meta-Learning for Biased Regularization and Fine Tuning

  • Giulia Denevi
  • Massimiliano Pontil
  • Carlo Ciliberto

Biased regularization and fine tuning are two recent meta-learning approaches. They have been shown to be effective to tackle distributions of tasks, in which the tasks’ target vectors are all close to a common meta-parameter vector. However, these methods may perform poorly on heterogeneous environments of tasks, where the complexity of the tasks’ distribution cannot be captured by a single meta- parameter vector. We address this limitation by conditional meta-learning, inferring a conditioning function mapping task’s side information into a meta-parameter vector that is appropriate for that task at hand. We characterize properties of the environment under which the conditional approach brings a substantial advantage over standard meta-learning and we highlight examples of environments, such as those with multiple clusters, satisfying these properties. We then propose a convex meta-algorithm providing a comparable advantage also in practice. Numerical experiments confirm our theoretical findings.

PDF Details

ICML Conference 2019 Conference Paper

Learning-to-Learn Stochastic Gradient Descent with Biased Regularization

  • Giulia Denevi
  • Carlo Ciliberto
  • Riccardo Grazzi
  • Massimiliano Pontil

We study the problem of learning-to-learn: infer- ring a learning algorithm that works well on a family of tasks sampled from an unknown distribution. As class of algorithms we consider Stochastic Gradient Descent (SGD) on the true risk regularized by the square euclidean distance from a bias vector. We present an average excess risk bound for such a learning algorithm that quantifies the potential benefit of using a bias vector with respect to the unbiased case. We then propose a novel meta-algorithm to estimate the bias term online from a sequence of observed tasks. The small memory footprint and low time complexity of our approach makes it appealing in practice while our theoretical analysis provides guarantees on the generalization properties of the meta-algorithm on new tasks. A key feature of our results is that, when the number of tasks grows and their vari- ance is relatively small, our learning-to-learn approach has a significant advantage over learning each task in isolation by standard SGD without a bias term. Numerical experiments demonstrate the effectiveness of our approach in practice.

Details

ICML Conference 2019 Conference Paper

Leveraging Low-Rank Relations Between Surrogate Tasks in Structured Prediction

  • Giulia Luise
  • Dimitrios Stamos
  • Massimiliano Pontil
  • Carlo Ciliberto

We study the interplay between surrogate methods for structured prediction and techniques from multitask learning designed to leverage relationships between surrogate outputs. We propose an efficient algorithm based on trace norm regularization which, differently from previous methods, does not require explicit knowledge of the coding/decoding functions of the surrogate framework. As a result, our algorithm can be applied to the broad class of problems in which the surrogate space is large or even infinite dimensional. We study excess risk bounds for trace norm regularized structured prediction proving the consistency and learning rates for our estimator. We also identify relevant regimes in which our approach can enjoy better generalization performance than previous methods. Numerical experiments on ranking problems indicate that enforcing low-rank relations among surrogate outputs may indeed provide a significant advantage in practice.

Details

NeurIPS Conference 2019 Conference Paper

Localized Structured Prediction

  • Carlo Ciliberto
  • Francis Bach
  • Alessandro Rudi

Key to structured prediction is exploiting the problem's structure to simplify the learning process. A major challenge arises when data exhibit a local structure (i. e. , are made ``by parts'') that can be leveraged to better approximate the relation between (parts of) the input and (parts of) the output. Recent literature on signal processing, and in particular computer vision, shows that capturing these aspects is indeed essential to achieve state-of-the-art performance. However, in this context algorithms are typically derived on a case-by-case basis. In this work we propose the first theoretical framework to deal with part-based data from a general perspective and study a novel method within the setting of statistical learning theory. Our analysis is novel in that it explicitly quantifies the benefits of leveraging the part-based structure of a problem on the learning rates of the proposed estimator.

PDF Details

NeurIPS Conference 2019 Conference Paper

Online-Within-Online Meta-Learning

  • Giulia Denevi
  • Dimitris Stamos
  • Carlo Ciliberto
  • Massimiliano Pontil

We study the problem of learning a series of tasks in a fully online Meta-Learning setting. The goal is to exploit similarities among the tasks to incrementally adapt an inner online algorithm in order to incur a low averaged cumulative error over the tasks. We focus on a family of inner algorithms based on a parametrized variant of online Mirror Descent. The inner algorithm is incrementally adapted by an online Mirror Descent meta-algorithm using the corresponding within-task minimum regularized empirical risk as the meta-loss. In order to keep the process fully online, we approximate the meta-subgradients by the online inner algorithm. An upper bound on the approximation error allows us to derive a cumulative error bound for the proposed method. Our analysis can also be converted to the statistical setting by online-to-batch arguments. We instantiate two examples of the framework in which the meta-parameter is either a common bias vector or feature map. Finally, preliminary numerical experiments confirm our theoretical findings.

PDF Details

ICML Conference 2019 Conference Paper

Random Expert Distillation: Imitation Learning via Expert Policy Support Estimation

  • Ruohan Wang
  • Carlo Ciliberto
  • Pierluigi Vito Amadori
  • Yiannis Demiris

We consider the problem of imitation learning from a finite set of expert trajectories, without access to reinforcement signals. The classical approach of extracting the expert’s reward function via inverse reinforcement learning, followed by reinforcement learning is indirect and may be computationally expensive. Recent generative adversarial methods based on matching the policy distribution between the expert and the agent could be unstable during training. We propose a new framework for imitation learning by estimating the support of the expert policy to compute a fixed reward function, which allows us to re-frame imitation learning within the standard reinforcement learning setting. We demonstrate the efficacy of our reward function on both discrete and continuous domains, achieving comparable or better performance than the state of the art under different reinforcement learning algorithms.

Details

NeurIPS Conference 2019 Conference Paper

Sinkhorn Barycenters with Free Support via Frank-Wolfe Algorithm

  • Giulia Luise
  • Saverio Salzo
  • Massimiliano Pontil
  • Carlo Ciliberto

We present a novel algorithm to estimate the barycenter of arbitrary probability distributions with respect to the Sinkhorn divergence. Based on a Frank-Wolfe optimization strategy, our approach proceeds by populating the support of the barycenter incrementally, without requiring any pre-allocation. We consider discrete as well as continuous distributions, proving convergence rates of the proposed algorithm in both settings. Key elements of our analysis are a new result showing that the Sinkhorn divergence on compact domains has Lipschitz continuous gradient with respect to the Total Variation and a characterization of the sample complexity of Sinkhorn potentials. Experiments validate the effectiveness of our method in practice.

PDF Details

NeurIPS Conference 2018 Conference Paper

Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance

  • Giulia Luise
  • Alessandro Rudi
  • Massimiliano Pontil
  • Carlo Ciliberto

Applications of optimal transport have recently gained remarkable attention as a result of the computational advantages of entropic regularization. However, in most situations the Sinkhorn approximation to the Wasserstein distance is replaced by a regularized version that is less accurate but easy to differentiate. In this work we characterize the differential properties of the original Sinkhorn approximation, proving that it enjoys the same smoothness as its regularized version and we explicitly provide an efficient algorithm to compute its gradient. We show that this result benefits both theory and applications: on one hand, high order smoothness confers statistical guarantees to learning with Wasserstein approximations. On the other hand, the gradient formula allows to efficiently solve learning and optimization problems in practice. Promising preliminary experiments complement our analysis.

PDF Details

UAI Conference 2018 Conference Paper

Incremental Learning-to-Learn with Statistical Guarantees

  • Giulia Denevi
  • Carlo Ciliberto
  • Dimitrios Stamos
  • Massimiliano Pontil

In learning-to-learn the goal is to infer a learning algorithm that works well on a class of tasks sampled from an unknown metadistribution. In contrast to previous work on batch learning-to-learn, we consider a scenario where tasks are presented sequentially and the algorithm needs to adapt incrementally to improve its performance on future tasks. Key to this setting is for the algorithm to rapidly incorporate new observations into the model as they arrive, without keeping them in memory. We focus on the case where the underlying algorithm is Ridge Regression parametrised by a symmetric positive semidefinite matrix. We propose to learn this matrix by applying a stochastic strategy to minimize the empirical error incurred by Ridge Regression on future tasks sampled from the meta-distribution. We study the statistical properties of the proposed algorithm and prove non-asymptotic bounds on its excess transfer risk, that is, the generalization performance on new tasks from the same meta-distribution. We compare our online learning-to-learn approach with a state-ofthe-art batch method, both theoretically and empirically.

Details

NeurIPS Conference 2018 Conference Paper

Learning To Learn Around A Common Mean

  • Giulia Denevi
  • Carlo Ciliberto
  • Dimitris Stamos
  • Massimiliano Pontil

The problem of learning-to-learn (LTL) or meta-learning is gaining increasing attention due to recent empirical evidence of its effectiveness in applications. The goal addressed in LTL is to select an algorithm that works well on tasks sampled from a meta-distribution. In this work, we consider the family of algorithms given by a variant of Ridge Regression, in which the regularizer is the square distance to an unknown mean vector. We show that, in this setting, the LTL problem can be reformulated as a Least Squares (LS) problem and we exploit a novel meta- algorithm to efficiently solve it. At each iteration the meta-algorithm processes only one dataset. Specifically, it firstly estimates the stochastic LS objective function, by splitting this dataset into two subsets used to train and test the inner algorithm, respectively. Secondly, it performs a stochastic gradient step with the estimated value. Under specific assumptions, we present a bound for the generalization error of our meta-algorithm, which suggests the right splitting parameter to choose. When the hyper-parameters of the problem are fixed, this bound is consistent as the number of tasks grows, even if the sample size is kept constant. Preliminary experiments confirm our theoretical findings, highlighting the advantage of our approach, with respect to independent task learning.

PDF Details

NeurIPS Conference 2018 Conference Paper

Manifold Structured Prediction

  • Alessandro Rudi
  • Carlo Ciliberto
  • GianMaria Marconi
  • Lorenzo Rosasco

Structured prediction provides a general framework to deal with supervised problems where the outputs have semantically rich structure. While classical approaches consider finite, albeit potentially huge, output spaces, in this paper we discuss how structured prediction can be extended to a continuous scenario. Specifically, we study a structured prediction approach to manifold-valued regression. We characterize a class of problems for which the considered approach is statistically consistent and study how geometric optimization can be used to compute the corresponding estimator. Promising experimental results on both simulated and real data complete our study.

PDF Details

NeurIPS Conference 2017 Conference Paper

Consistent Multitask Learning with Nonlinear Output Relations

  • Carlo Ciliberto
  • Alessandro Rudi
  • Lorenzo Rosasco
  • Massimiliano Pontil

Key to multitask learning is exploiting the relationships between different tasks to improve prediction performance. Most previous methods have focused on the case where tasks relations can be modeled as linear operators and regularization approaches can be used successfully. However, in practice assuming the tasks to be linearly related is often restrictive, and allowing for nonlinear structures is a challenge. In this paper, we tackle this issue by casting the problem within the framework of structured prediction. Our main contribution is a novel algorithm for learning multiple tasks which are related by a system of nonlinear equations that their joint outputs need to satisfy. We show that our algorithm can be efficiently implemented and study its generalization properties, proving universal consistency and learning rates. Our theoretical analysis highlights the benefits of non-linear multitask learning over learning the tasks independently. Encouraging experimental results show the benefits of the proposed method in practice.

PDF Details

ICRA Conference 2017 Conference Paper

Incremental robot learning of new objects with fixed update time

  • Raffaello Camoriano
  • Giulia Pasquale
  • Carlo Ciliberto
  • Lorenzo Natale
  • Lorenzo Rosasco
  • Giorgio Metta

We consider object recognition in the context of lifelong learning, where a robotic agent learns to discriminate between a growing number of object classes as it accumulates experience about the environment. We propose an incremental variant of the Regularized Least Squares for Classification (RLSC) algorithm, and exploit its structure to seamlessly add new classes to the learned model. The presented algorithm addresses the problem of having an unbalanced proportion of training examples per class, which occurs when new objects are presented to the system for the first time. We evaluate our algorithm on both a machine learning benchmark dataset and two challenging object recognition tasks in a robotic setting. Empirical evidence shows that our approach achieves comparable or higher classification performance than its batch counterpart when classes are unbalanced, while being significantly faster.

Details

NeurIPS Conference 2016 Conference Paper

A Consistent Regularization Approach for Structured Prediction

  • Carlo Ciliberto
  • Lorenzo Rosasco
  • Alessandro Rudi

We propose and analyze a regularization approach for structured prediction problems. We characterize a large class of loss functions that allows to naturally embed structured outputs in a linear space. We exploit this fact to design learning algorithms using a surrogate loss approach and regularization techniques. We prove universal consistency and finite sample bounds characterizing the generalization properties of the proposed method. Experimental results are provided to demonstrate the practical usefulness of the proposed approach.

PDF Details

IROS Conference 2016 Conference Paper

Object identification from few examples by improving the invariance of a Deep Convolutional Neural Network

  • Giulia Pasquale
  • Carlo Ciliberto
  • Lorenzo Rosasco
  • Lorenzo Natale

The development of reliable and robust visual recognition systems is a main challenge towards the deployment of autonomous robotic agents in unconstrained environments. Learning to recognize objects requires image representations that are discriminative to relevant information while being invariant to nuisances, such as scaling, rotations, light and background changes, and so forth. Deep Convolutional Neural Networks can learn such representations from large web-collected image datasets and a natural question is how these systems can be best adapted to the robotics context where little supervision is often available. In this work, we investigate different training strategies for deep architectures on a new dataset collected in a real-world robotic setting. In particular we show how deep networks can be tuned to improve invariance and discriminability properties and perform object identification tasks with minimal supervision.

Details

ICML Conference 2015 Conference Paper

Convex Learning of Multiple Tasks and their Structure

  • Carlo Ciliberto
  • Youssef Mroueh
  • Tomaso A. Poggio
  • Lorenzo Rosasco

Reducing the amount of human supervision is a key problem in machine learning and a natural approach is that of exploiting the relations (structure) among different tasks. This is the idea at the core of multi-task learning. In this context a fundamental question is how to incorporate the tasks structure in the learning problem. We tackle this question by studying a general computational framework that allows to encode a-priori knowledge of the tasks structure in the form of a convex penalty; in this setting a variety of previously proposed methods can be recovered as special cases, including linear and non-linear approaches. Within this framework, we show that tasks and their structure can be efficiently learned considering a convex optimization problem that can be approached by means of block coordinate methods such as alternating minimization and for which we prove convergence to the global minimum.

Details

IROS Conference 2014 Conference Paper

Exploiting global force torque measurements for local compliance estimation in tactile arrays

  • Carlo Ciliberto
  • Luca Fiorio
  • Marco Maggiali
  • Lorenzo Natale
  • Lorenzo Rosasco
  • Giorgio Metta
  • Giulio Sandini
  • Francesco Nori

In this paper we tackle the problem of estimating the local compliance of tactile arrays exploiting global measurements from a single force and torque sensor. The proposed procedure exploits a transformation matrix (describing the relative position between the local tactile elements and the global force/torque measurements) to define a linear regression problem on the unknown local stiffness. Experiments have been conducted on the foot of the iCub robot, sensorized with a single force/torque sensor and a tactile array of 250 tactile elements (taxels) on the foot sole. Results show that a simple calibration procedure can be employed to estimate the stiffness parameters of virtual springs over a tactile array and to use these model to predict normal forces exerted on the array based only on the tactile feedback. Leveraging on previous works [1] the proposed procedure does not necessarily need a-priori information on the transformation matrix of the taxels which can be directly estimated from available measurements.

Details

IROS Conference 2013 Conference Paper

On the impact of learning hierarchical representations for visual recognition in robotics

  • Carlo Ciliberto
  • Sean Ryan Fanello
  • Matteo Santoro
  • Lorenzo Natale
  • Giorgio Metta
  • Lorenzo Rosasco

Recent developments in learning sophisticated, hierarchical image representations have led to remarkable progress in the context of visual recognition. While these methods are becoming standard in modern computer vision systems, they are rarely adopted in robotics. The question arises of whether solutions, which have been primarily developed for image retrieval, can perform well in more dynamic and unstructured scenarios. In this paper we tackle this question performing an extensive evaluation of state of the art methods for visual recognition on a iCub robot. We consider the problem of classifying 15 different objects shown by a human demonstrator in a challenging Human-Robot Interaction scenario. The classification performance of hierarchical learning approaches are shown to outperform benchmark solutions based on local descriptors and template matching. Our results show that hierarchical learning systems are computationally efficient and can be used for real-time training and recognition of objects.

Details

ICRA Conference 2013 Conference Paper

Weakly supervised strategies for natural object recognition in robotics

  • Sean Ryan Fanello
  • Carlo Ciliberto
  • Lorenzo Natale
  • Giorgio Metta

The paper aims at building a computer vision system for automatic image labeling in robotics scenarios. We show that the weak supervision provided by a human demonstrator, through the exploitation of the independent motion, enables a realistic Human-Robot Interaction (HRI) and achieves an automatic image labeling. We start by reviewing the underlying principles of our previous method for egomotion compensation [1], then we extend our approach removing the dependency on a known kinematics in order to provide a general method for a wide range of devices. From sparse salient features we predict the egomotion of the camera through a heteroscedastic learning method. Subsequently we use an object recognition framework for testing the automatic image labeling process: we rely on the State of the Art method from Yang et al. [2], employing local features augmented through a sparse coding stage and classified with linear SVMs. The application has been implemented and validated on the iCub humanoid robot and experiments are presented to show the effectiveness of the proposed approach. The contribution of the paper is twofold: first we overcome the dependency on the kinematics in the independent motion detection method, secondly we present a practical application for automatic image labeling through a natural HRI.

Details

IROS Conference 2012 Conference Paper

A heteroscedastic approach to independent motion detection for actuated visual sensors

  • Carlo Ciliberto
  • Sean Ryan Fanello
  • Lorenzo Natale
  • Giorgio Metta

We present an original method for independent motion detection in dynamic scenes. The algorithm is designed for robotics real-time applications and it overcomes the short-comings of current approaches for the egomotion estimation in presence of many outliers, occlusions and cluttered background. The method relies on a stereo system which performs the reprojection of a sparse set of features following the camera displacement. We assume that noisy prior knowledge of the motion is available (i. e. a robot's kinematic model). Since this estimation leads to a heteroscedastic regression problem due to input-dependent noise, we employ a simple, but computationally efficient approach in order to accurately determine the latent egomotion subspace spanned by the Degrees of Freedom (DOFs) of the robot. The algorithm has been implemented and validated on the iCub humanoid robot. Qualitative and quantitative experiments are presented to show the effectiveness of the proposed approach. The contribution of the paper is a modular framework for independent motion detection naturally extendable to any architecture featuring a visual sensor that can be directly controllable.

Details

IROS Conference 2011 Conference Paper

Online multiple instance learning applied to hand detection in a humanoid robot

  • Carlo Ciliberto
  • Fabrizio Smeraldi
  • Lorenzo Natale
  • Giorgio Metta

We propose an algorithm for the visual detection and localisation of the hand of a humanoid robot. This algorithm imposes low requirements on the type of supervision required to achieve good performance. In particular the system performs feature selection and adaptation using images that are only labelled as containing the hand or not, without any explicit segmentation. Our algorithm is an online variant of Multiple Instance Learning based on boosting. Experiments in real-world conditions on the iCub humanoid robot confirm that the algorithm can learn the visual appearance of the hand, reaching an accuracy comparable with its off-line version. This remains true when supervision is generated by the robot itself in a completely autonomous fashion. Algorithms with weak supervision requirements like the one we describe are useful for autonomous robots that learn and adapt online to a changing environment. The algorithm is not hand-specific and could be easily applied to wide range of problems involving visual recognition of generic objects.

Details

IROS Conference 2011 Conference Paper

Reexamining Lucas-Kanade method for real-time independent motion detection: Application to the iCub humanoid robot

  • Carlo Ciliberto
  • Ugo Pattacini
  • Lorenzo Natale
  • Francesco Nori
  • Giorgio Metta

Visual motion is a simple yet powerful cue widely used by biological systems to improve their perception and adaptation to the environment. Examples of tasks that greatly benefit from the ability to detect movement are object segmentation, 3D scene reconstruction and control of attention. In computer vision several algorithms for computing visual motion and optic flow exist. However their application in robotics is not straightforward as in these platforms visual motion is often dominated by (self) motion produced by the movement of the robot (egomotion) making it difficult to disambiguate between motion induced by the scene dynamics or by the own actions of the robot. Independent motion detection is an active field in computer vision and robotics, however approaches in this area typically require that some models of both the environment and the robot visual system are available and are hardly suitable for real-time control. In this paper we describe the motionCUT, a derivation of the Lucas-Kanade optical flow algorithm that allows detecting moving objects, irrespectively of the egomotion produced by the robot. Our method is purely visual and does not require information other than the images coming from the cameras. As such it can be easily adapted to any robotic platform. The system was tested on a stereo tracking task on the iCub humanoid robot, demonstrating that the algorithm performs well and can easily execute in real-time.

Details
v2026.09.13