Eric Eaton

TMLR Journal 2026 Journal Article

Iterative Compositional Data Generation for Robot Control

• Anh-Quan Pham
• Marcel Hussing
• Shubhankar P. Patankar
• Danielle Bassett
• Jorge Mendez-Mendez
• Eric Eaton

Collecting robotic manipulation data is expensive, making it impractical to acquire demonstrations for the combinatorially large space of tasks that arise in multi-object, multi-robot, and multi-environment settings. While recent generative models can synthesize useful data for individual tasks, they do not exploit the compositional structure of robotic domains and struggle to generalize to unseen task combinations. We propose a semantic compositional diffusion transformer that factorizes transitions into robot-, object-, obstacle-, and objective-specific components and learns their interactions through attention. Once trained on a limited subset of tasks, we show that our model can zero-shot generate high-quality transitions from which we can learn control policies for unseen task combinations. Then, we introduce an iterative self-improvement procedure in which synthetic data is validated via offline reinforcement learning and incorporated into subsequent training rounds. Our approach substantially improves zero-shot performance over monolithic and hard-coded compositional baselines, ultimately solving nearly all held-out tasks and demonstrating the emergence of meaningful compositional structure in the learned representations.

ICLR Conference 2025 Conference Paper

Articulate-Anything: Automatic Modeling of Articulated Objects via a Vision-Language Foundation Model

• Long Le
• Jason Xie
• William Liang
• Hung-Ju Wang
• Yue Yang
• Yecheng Jason Ma 0001
• Kyle Vedder
• Arjun Krishna

Interactive 3D simulated objects are crucial in AR/VR, animations, and robotics, driving immersive experiences and advanced automation. However, creating these articulated objects requires extensive human effort and expertise, limiting their broader applications. To overcome this challenge, we present Articulate-Anything, a system that automates the articulation of diverse, complex objects from many input modalities, including text, images, and videos. Articulate-Anything leverages vision-language models (VLMs) to generate code that can be compiled into an interactable digital twin for use in standard 3D simulators. Our system exploits existing 3D asset datasets via a mesh retrieval mechanism, along with an actor-critic system that iteratively proposes, evaluates, and refines solutions for articulating the objects, self-correcting errors to achieve a robust out- come. Qualitative evaluations demonstrate Articulate-Anything's capability to articulate complex and even ambiguous object affordances by leveraging rich grounded inputs. In extensive quantitative experiments on the standard PartNet-Mobility dataset, Articulate-Anything substantially outperforms prior work, increasing the success rate from 8.7-11.6\% to 75\% and setting a new bar for state-of-art performance. We further showcase the utility of our generated assets by using them to train robotic policies for fine-grained manipulation tasks that go beyond basic pick and place.

AAAI Conference 2025 Conference Paper

Assessing Modality Bias in Video Question Answering Benchmarks with Multimodal Large Language Models

• Jean Park
• Kuk Jin Jang
• Basam Alasaly
• Sriharsha Mopidevi
• Andrew Zolensky
• Eric Eaton
• Insup Lee
• Kevin Johnson

Multimodal large language models (MLLMs) can simultaneously process visual, textual, and auditory data, capturing insights that complement human analysis. However, existing video question-answering (VidQA) benchmarks and datasets often exhibit a bias toward a single modality, despite the goal of requiring advanced reasoning skills that integrate diverse modalities to answer the queries. In this work, we introduce the modality importance score (MIS) to identify such bias. It is designed to assess which modality embeds the necessary information to answer the question. Additionally, we propose an innovative method using state-of-the-art MLLMs to estimate the modality importance, which can serve as a proxy for human judgments of modality perception. With this MIS, we demonstrate the presence of unimodal bias and the scarcity of genuinely multimodal questions in existing datasets. We further validate the modality importance score with multiple ablation studies to evaluate the performance of MLLMs on permuted feature sets. Our results indicate that current models do not effectively integrate information due to modality imbalance in existing datasets. Our proposed MLLM-derived MIS can guide the curation of modality-balanced datasets that advance multimodal learning and enhance MLLMs' capabilities to understand and utilize synergistic relations across modalities.

NeurIPS Conference 2025 Conference Paper

FORLA: Federated Object-Centric Representation Learning with Slot Attention

• Guiqiu Liao
• Matjaz Jogan
• Eric Eaton
• Daniel Hashimoto

Learning efficient visual representations across heterogeneous unlabeled datasets remains a central challenge in federated learning. Effective federated representations require features that are jointly informative across clients while disentangling domain-specific factors without supervision. We introduce FORLA, a novel framework for federated object-centric representation learning and feature adaptation across clients using unsupervised slot attention. At the core of our method is a shared feature adapter, trained collaboratively across clients to adapt features from foundation models, and a shared slot attention module that learns to reconstruct the adapted features. To optimize this adapter, we design a two-branch student–teacher architecture. In each client, a student decoder learns to reconstruct full features from foundation models, while a teacher decoder reconstructs their adapted, low-dimensional counterpart. The shared slot attention module bridges cross-domain learning by aligning object-level representations across clients. Experiments in multiple real-world datasets show that our framework not only outperforms centralized baselines on object discovery but also learns a compact, universal representation that generalizes well across domains. This work highlights federated slot attention as an effective tool for scalable, unsupervised visual representation learning from cross-domain data with distributed concepts.

ICML Conference 2025 Conference Paper

Intersectional Fairness in Reinforcement Learning with Large State and Constraint Spaces

• Eric Eaton
• Marcel Hussing
• Michael J. Kearns
• Aaron Roth 0001
• Sikata Bela Sengupta
• Jessica Sorrell

In traditional reinforcement learning (RL), the learner aims to solve a single objective optimization problem: find the policy that maximizes expected reward. However, in many real-world settings, it is important to optimize over multiple objectives simultaneously. For example, when we are interested in fairness, states might have feature annotations corresponding to multiple (intersecting) demographic groups to whom reward accrues, and our goal might be to maximize the reward of the group receiving the minimal reward. In this work, we consider a multi-objective optimization problem in which each objective is defined by a state-based reweighting of a single scalar reward function. This generalizes the problem of maximizing the reward of the minimum reward group. We provide oracle-efficient algorithms to solve these multi-objective RL problems even when the number of objectives is very large — for tabular MDPs, as well as for large MDPs when the group functions have additional structure. The contribution of this paper is that we are able to solve this class of multi-objective RL problems with a possibly exponentially large class of constraints over intersecting groups in both tabular and large state space MDPs in an oracle-efficient manner. Finally, we experimentally validate our theoretical results and demonstrate applications on a preferential attachment graph MDP.

ICLR Conference 2025 Conference Paper

MAD-TD: Model-Augmented Data stabilizes High Update Ratio RL

• Claas Voelcker
• Marcel Hussing
• Eric Eaton
• Amir Massoud Farahmand
• Igor Gilitschenski

Building deep reinforcement learning (RL) agents that find a good policy with few samples has proven notoriously challenging. To achieve sample efficiency, recent work has explored updating neural networks with large numbers of gradient steps for every new sample. While such high update-to-data (UTD) ratios have shown strong empirical performance, they also introduce instability to the training process. Previous approaches need to rely on periodic neural network parameter resets to address this instability, but restarting the training process is infeasible in many real-world applications and requires tuning the resetting interval. In this paper, we focus on one of the core difficulties of stable training with limited samples: the inability of learned value functions to generalize to unobserved on-policy actions. We mitigate this issue directly by augmenting the off-policy RL training process with a small amount of data generated from a learned world model. Our method, Model-Augmented Data for TD Learning (MAD-TD) uses small amounts of generated data to stabilize high UTD training and achieve competitive performance on the most challenging tasks in the DeepMind control suite. Our experiments further highlight the importance of employing a good model to generate data, MAD-TD's ability to combat value overestimation, and its practical stability gains for continued learning.

ICLR Conference 2025 Conference Paper

Neural Eulerian Scene Flow Fields

• Kyle Vedder
• Neehar Peri
• Ishan Khatri
• Siyi Li
• Eric Eaton
• Mehmet Kemal Kocamaz
• Yue Wang
• Zhiding Yu

We reframe scene flow as the task of estimating a continuous space-time ordinary differential equation (ODE) that describes motion for an entire observation sequence, represented with a neural prior. Our method, EulerFlow, optimizes this neural prior estimate against several multi-observation reconstruction objectives, enabling high quality scene flow estimation via self-supervision on real-world data. EulerFlow works out-of-the-box without tuning across multiple domains, including large-scale autonomous driving scenes and dynamic tabletop settings. Remarkably, EulerFlow produces high quality flow estimates on small, fast moving objects like birds and tennis balls, and exhibits emergent 3D point tracking behavior by solving its estimated ODE over long-time horizons. On the Argoverse 2 2024 Scene Flow Challenge, EulerFlow outperforms all prior art, surpassing the next-best unsupervised method by more than 2.5 times, and even exceeding the next-best supervised method by over 10%. See https://vedder.io/eulerflow for interactive visuals.

ICRA Conference 2024 Conference Paper

A Metacognitive Approach to Out-of-Distribution Detection for Segmentation

• Meghna Gummadi
• David Kent 0001
• Karl Schmeckpeper
• Eric Eaton

Despite outstanding semantic scene segmentation in closed-worlds, deep neural networks segment novel instances poorly, which is required for autonomous agents acting in an open world. To improve out-of-distribution (OOD) detection for segmentation, we introduce a metacognitive approach in the form of a lightweight module that leverages entropy measures, segmentation predictions, and spatial context to characterize the segmentation model’s uncertainty and detect pixel-wise OOD data in real-time. Additionally, our approach incorporates a novel method of generating synthetic OOD data in context with in-distribution data, which we use to fine-tune existing segmentation models with maximum entropy training. This further improves the metacognitive module’s performance without requiring access to OOD data while enabling compatibility with established pre-trained models. Our resulting approach can reliably detect OOD instances in a scene, as shown by state-of-the-art performance on OOD detection for semantic segmentation benchmarks.

RLJ Journal 2024 Journal Article

Dissecting Deep RL with High Update Ratios: Combatting Value Divergence

• Marcel Hussing
• Claas A Voelcker
• Igor Gilitschenski
• Amir-massoud Farahmand
• Eric Eaton

We show that deep reinforcement learning algorithms can retain their ability to learn without resetting network parameters in settings where the number of gradient updates greatly exceeds the number of environment samples by combatting value function divergence. Under large update-to-data ratios, a recent study by Nikishin et al. (2022) suggested the emergence of a primacy bias, in which agents overfit early interactions and downplay later experience, impairing their ability to learn. In this work, we investigate the phenomena leading to the primacy bias. We inspect the early stages of training that were conjectured to cause the failure to learn and find that one fundamental challenge is a long-standing acquaintance: value function divergence. Overinflated Q-values are found not only on out-of-distribution but also in-distribution data and can be linked to overestimation on unseen action prediction propelled by optimizer momentum. We employ a simple unit-ball normalization that enables learning under large update ratios, show its efficacy on the widely used dm_control suite, and obtain strong performance on the challenging dog tasks, competitive with model-based approaches. Our results question, in parts, the prior explanation for sub-optimal learning due to overfitting early data.

RLC Conference 2024 Conference Paper

Dissecting Deep RL with High Update Ratios: Combatting Value Divergence

• Marcel Hussing
• Claas A Voelcker
• Igor Gilitschenski
• Amir-massoud Farahm
• Eric Eaton

We show that deep reinforcement learning algorithms can retain their ability to learn without resetting network parameters in settings where the number of gradient updates greatly exceeds the number of environment samples by combatting value function divergence. Under large update-to-data ratios, a recent study by Nikishin et al. (2022) suggested the emergence of a primacy bias, in which agents overfit early interactions and downplay later experience, impairing their ability to learn. In this work, we investigate the phenomena leading to the primacy bias. We inspect the early stages of training that were conjectured to cause the failure to learn and find that one fundamental challenge is a long-standing acquaintance: value function divergence. Overinflated Q-values are found not only on out-of-distribution but also in-distribution data and can be linked to overestimation on unseen action prediction propelled by optimizer momentum. We employ a simple unit-ball normalization that enables learning under large update ratios, show its efficacy on the widely used dm_control suite, and obtain strong performance on the challenging dog tasks, competitive with model-based approaches. Our results question, in parts, the prior explanation for sub-optimal learning due to overfitting early data.

RLC Conference 2024 Conference Paper

Robotic Manipulation Datasets for Offline Compositional Reinforcement Learning

• Marcel Hussing
• Jorge Mendez-Mendez
• Anisha Singrodia
• Cassandra Kent
• Eric Eaton

Offline reinforcement learning (RL) is a promising direction that allows RL agents to pre-train on large datasets, avoiding the recurrence of expensive data collection. To advance the field, it is crucial to generate large-scale datasets. Compositional RL is particularly appealing for generating such large datasets, since 1) it permits creating many tasks from few components, 2) the task structure may enable trained agents to solve new tasks by combining relevant learned components, and 3) the compositional dimensions provide a notion of task relatedness. This paper provides four offline RL datasets for simulated robotic manipulation created using the $256$ tasks from CompoSuite (Mendez et al. , 2022). Each dataset is collected from an agent with a different degree of performance, and consists of $256$ million transitions. We provide training and evaluation settings for assessing an agent's ability to learn compositional task policies. Our benchmarking experiments show that current offline RL methods can learn the training tasks to some extent and that compositional methods outperform non-compositional methods. Yet, current methods are unable to extract the compositional structure to generalize to unseen tasks highlighting a need for future research in offline compositional RL.

RLJ Journal 2024 Journal Article

Robotic Manipulation Datasets for Offline Compositional Reinforcement Learning

• Marcel Hussing
• Jorge Mendez-Mendez
• Anisha Singrodia
• Cassandra Kent
• Eric Eaton

Offline reinforcement learning (RL) is a promising direction that allows RL agents to pre-train on large datasets, avoiding the recurrence of expensive data collection. To advance the field, it is crucial to generate large-scale datasets. Compositional RL is particularly appealing for generating such large datasets, since 1) it permits creating many tasks from few components, 2) the task structure may enable trained agents to solve new tasks by combining relevant learned components, and 3) the compositional dimensions provide a notion of task relatedness. This paper provides four offline RL datasets for simulated robotic manipulation created using the $256$ tasks from CompoSuite (Mendez et al., 2022). Each dataset is collected from an agent with a different degree of performance, and consists of $256$ million transitions. We provide training and evaluation settings for assessing an agent's ability to learn compositional task policies. Our benchmarking experiments show that current offline RL methods can learn the training tasks to some extent and that compositional methods outperform non-compositional methods. Yet, current methods are unable to extract the compositional structure to generalize to unseen tasks highlighting a need for future research in offline compositional RL.

ICLR Conference 2024 Conference Paper

ZeroFlow: Scalable Scene Flow via Distillation

• Kyle Vedder
• Neehar Peri
• Nathaniel Chodosh
• Ishan Khatri
• Eric Eaton
• Dinesh Jayaraman
• Yang Liu
• Deva Ramanan

Scene flow estimation is the task of describing the 3D motion field between temporally successive point clouds. State-of-the-art methods use strong priors and test-time optimization techniques, but require on the order of tens of seconds to process full-size point clouds, making them unusable as computer vision primitives for real-time applications such as open world object detection. Feedforward methods are considerably faster, running on the order of tens to hundreds of milliseconds for full-size point clouds, but require expensive human supervision. To address both limitations, we propose _Scene Flow via Distillation_, a simple, scalable distillation framework that uses a label-free optimization method to produce pseudo-labels to supervise a feedforward model. Our instantiation of this framework, _ZeroFlow_, achieves **state-of-the-art** performance on the _Argoverse 2 Self-Supervised Scene Flow Challenge_ while using zero human labels by simply training on large-scale, diverse unlabeled data. At test-time, ZeroFlow is over 1000$\times$ faster than label-free state-of-the-art optimization-based methods on full-size point clouds (34 FPS vs 0.028 FPS) and over 1000$\times$ cheaper to train on unlabeled data compared to the cost of human annotation (\\$394 vs ~\\$750,000). To facilitate further research, we will release our code, trained model weights, and high quality pseudo-labels for the Argoverse 2 and Waymo Open datasets.

ICRA Conference 2023 Conference Paper

CAROM Air - Vehicle Localization and Traffic Scene Reconstruction from Aerial Videos

• Duo Lu
• Eric Eaton
• Matt Weg
• Wei Wang
• Steven Como
• Jeffrey Wishart
• Hongbin Yu
• Yezhou Yang

Road traffic scene reconstruction from videos has been desirable by road safety regulators, city planners, researchers, and autonomous driving technology developers. However, it is expensive and unnecessary to cover every mile of the road with cameras mounted on the road infrastructure. This paper presents a method that can process aerial videos to vehicle trajectory data so that a traffic scene can be automatically reconstructed and accurately re-simulated using computers. On average, the vehicle localization error is about 0. 1 m to 0. 3 m using a consumer-grade drone flying at 120 meters. This project also compiles a dataset of 50 reconstructed road traffic scenes from about 100 hours of aerial videos to enable various downstream traffic analysis applications and facilitate further road traffic related research. The dataset is available at https://github.com/duolu/CAROM.

JMLR Journal 2023 Journal Article

Gap Minimization for Knowledge Sharing and Transfer

• Boyu Wang
• Jorge A. Mendez
• Changjian Shui
• Fan Zhou
• Di Wu
• Gezheng Xu
• Christian Gagné
• Eric Eaton

Learning from multiple related tasks by knowledge sharing and transfer has become increasingly relevant over the last two decades. In order to successfully transfer information from one task to another, it is critical to understand the similarities and differences between the domains. In this paper, we introduce the notion of performance gap, an intuitive and novel measure of the distance between learning tasks. Unlike existing measures which are used as tools to bound the difference of expected risks between tasks (e.g., $\mathcal{H}$-divergence or discrepancy distance), we theoretically show that the performance gap can be viewed as a data- and algorithm-dependent regularizer, which controls the model complexity and leads to finer guarantees. More importantly, it also provides new insights and motivates a novel principle for designing strategies for knowledge sharing and transfer: gap minimization. We instantiate this principle with two algorithms: 1. gapBoost, a novel and principled boosting algorithm that explicitly minimizes the performance gap between source and target domains for transfer learning; and 2. gapMTNN, a representation learning algorithm that reformulates gap minimization as semantic conditional matching for multitask learning. Our extensive evaluation on both transfer learning and multitask learning benchmark data sets shows that our methods outperform existing baselines. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2023. ( edit, beta )

TMLR Journal 2023 Journal Article

How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition

• Jorge A Mendez
• Eric Eaton

A major goal of artificial intelligence (AI) is to create an agent capable of acquiring a general understanding of the world. Such an agent would require the ability to continually accumulate and build upon its knowledge as it encounters new experiences. Lifelong or continual learning addresses this setting, whereby an agent faces a continual stream of problems and must strive to capture the knowledge necessary for solving each new task it encounters. If the agent is capable of accumulating knowledge in some form of compositional representation, it could then selectively reuse and combine relevant pieces of knowledge to construct novel solutions. Despite the intuitive appeal of this simple idea, the literatures on lifelong learning and compositional learning have proceeded largely separately. In an effort to promote developments that bridge between the two fields, this article surveys their respective research landscapes and discusses existing and future connections between them.

NeurIPS Conference 2023 Conference Paper

Replicable Reinforcement Learning

• Eric Eaton
• Marcel Hussing
• Michael Kearns
• Jessica Sorrell

The replicability crisis in the social, behavioral, and data sciences has led to the formulation of algorithm frameworks for replicability --- i. e. , a requirement that an algorithm produce identical outputs (with high probability) when run on two different samples from the same underlying distribution. While still in its infancy, provably replicable algorithms have been developed for many fundamental tasks in machine learning and statistics, including statistical query learning, the heavy hitters problem, and distribution testing. In this work we initiate the study of replicable reinforcement learning, providing a provably replicable algorithm for parallel value iteration, and a provably replicable version of R-Max in the episodic setting. These are the first formal replicability results for control problems, which present different challenges for replication than batch learning settings.

ICLR Conference 2022 Conference Paper

Modular Lifelong Reinforcement Learning via Neural Composition

• Jorge A. Mendez
• Harm van Seijen
• Eric Eaton

Humans commonly solve complex problems by decomposing them into easier subproblems and then combining the subproblem solutions. This type of compositional reasoning permits reuse of the subproblem solutions when tackling future tasks that share part of the underlying compositional structure. In a continual or lifelong reinforcement learning (RL) setting, this ability to decompose knowledge into reusable components would enable agents to quickly learn new RL tasks by leveraging accumulated compositional structures. We explore a particular form of composition based on neural modules and present a set of RL problems that intuitively admit compositional solutions. Empirically, we demonstrate that neural composition indeed captures the underlying structure of this space of problems. We further propose a compositional lifelong RL method that leverages accumulated neural components to accelerate the learning of future tasks while retaining performance on previous tasks via off-line RL over replayed experiences.

IROS Conference 2022 Conference Paper

Sparse PointPillars: Maintaining and Exploiting Input Sparsity to Improve Runtime on Embedded Systems

• Kyle Vedder
• Eric Eaton

Bird's Eye View (BEV) is a popular representation for processing 3D point clouds, and by its nature is fundamentally sparse. Motivated by the computational limitations of mobile robot platforms, we create a fast, high-performance BEV 3D object detector that maintains and exploits this input sparsity to decrease runtimes over non-sparse baselines and avoids the tradeoff between pseudoimage area and runtime. We present results on KITTI, a canonical 3D detection dataset, and Matterport-Chair, a novel Matterport3D-derived chair detection dataset from scenes in real furnished homes. We evaluate runtime characteristics using a desktop GPU, an embedded ML accelerator, and a robot CPU, demonstrating that our method results in significant detection speedups (2 × or more) for embedded systems with only a modest decrease in detection quality. Our work represents a new approach for practitioners to optimize models for embedded systems by maintaining and exploiting input sparsity throughout their entire pipeline to reduce runtime and resource usage while preserving detection performance. All models, weights, experimental configurations, and datasets used are publicly available 1 1 https://vedder.io/sparse_point_pillars.

ICLR Conference 2021 Conference Paper

Lifelong Learning of Compositional Structures

• Jorge A. Mendez
• Eric Eaton

A hallmark of human intelligence is the ability to construct self-contained chunks of knowledge and adequately reuse them in novel combinations for solving different yet structurally related problems. Learning such compositional structures has been a significant challenge for artificial systems, due to the combinatorial nature of the underlying search problem. To date, research into compositional learning has largely proceeded separately from work on lifelong or continual learning. We integrate these two lines of work to present a general-purpose framework for lifelong learning of compositional structures that can be used for solving a stream of related tasks. Our framework separates the learning process into two broad stages: learning how to best combine existing components in order to assimilate a novel problem, and learning how to adapt the set of existing components to accommodate the new problem. This separation explicitly handles the trade-off between the stability required to remember how to solve earlier tasks and the flexibility required to solve new tasks, as we show empirically in an extensive evaluation.

ICML Conference 2021 Conference Paper

Sharing Less is More: Lifelong Learning in Deep Networks with Selective Layer Transfer

• Seungwon Lee
• Sima Behpour
• Eric Eaton

Effective lifelong learning across diverse tasks requires the transfer of diverse knowledge, yet transferring irrelevant knowledge may lead to interference and catastrophic forgetting. In deep networks, transferring the appropriate granularity of knowledge is as important as the transfer mechanism, and must be driven by the relationships among tasks. We first show that the lifelong learning performance of several current deep learning architectures can be significantly improved by transfer at the appropriate layers. We then develop an expectation-maximization (EM) method to automatically select the appropriate transfer configuration and optimize the task network weights. This EM-based selective transfer is highly effective, balancing transfer performance on all tasks with avoiding catastrophic forgetting, as demonstrated on three algorithms in several lifelong object classification scenarios.

NeurIPS Conference 2020 Conference Paper

Lifelong Policy Gradient Learning of Factored Policies for Faster Training Without Forgetting

• Jorge Mendez
• Boyu Wang
• Eric Eaton

Policy gradient methods have shown success in learning control policies for high-dimensional dynamical systems. Their biggest downside is the amount of exploration they require before yielding high-performing policies. In a lifelong learning setting, in which an agent is faced with multiple consecutive tasks over its lifetime, reusing information from previously seen tasks can substantially accelerate the learning of new tasks. We provide a novel method for lifelong policy gradient learning that trains lifelong function approximators directly via policy gradients, allowing the agent to benefit from accumulated knowledge throughout the entire training process. We show empirically that our algorithm learns faster and converges to better policies than single-task and lifelong learning baselines, and completely avoids catastrophic forgetting on a variety of challenging domains.

JAIR Journal 2020 Journal Article

Using Task Descriptions in Lifelong Machine Learning for Improved Performance and Zero-Shot Transfer

• Mohammad Rostami
• David Isele
• Eric Eaton

Knowledge transfer between tasks can improve the performance of learned models, but requires an accurate estimate of inter-task relationships to identify the relevant knowledge to transfer. These inter-task relationships are typically estimated based on training data for each task, which is inefficient in lifelong learning settings where the goal is to learn each consecutive task rapidly from as little data as possible. To reduce this burden, we develop a lifelong learning method based on coupled dictionary learning that utilizes high-level task descriptions to model inter-task relationships. We show that using task descriptors improves the performance of the learned task policies, providing both theoretical justification for the benefit and empirical demonstration of the improvement across a variety of learning problems. Given only the descriptor for a new task, the lifelong learner is also able to accurately predict a model for the new task through zero-shot learning using the coupled dictionary, eliminating the need to gather training data before addressing the task.

IJCAI Conference 2019 Conference Paper

Learning Shared Knowledge for Deep Lifelong Learning using Deconvolutional Networks

• Seungwon Lee
• James Stokes
• Eric Eaton

Current mechanisms for knowledge transfer in deep networks tend to either share the lower layers between tasks, or build upon representations trained on other tasks. However, existing work in non-deep multi-task and lifelong learning has shown success with using factorized representations of the model parameter space for transfer, permitting more flexible construction of task models. Inspired by this idea, we introduce a novel architecture for sharing latent factorized representations in convolutional neural networks (CNNs). The proposed approach, called a deconvolutional factorized CNN, uses a combination of deconvolutional factorization and tensor contraction to perform flexible transfer between tasks. Experiments on two computer vision data sets show that the DF-CNN achieves superior performance in challenging lifelong learning settings, resists catastrophic forgetting, and exhibits reverse transfer to improve previously learned tasks from subsequent experience without retraining.

NeurIPS Conference 2019 Conference Paper

Transfer Learning via Minimizing the Performance Gap Between Domains

• Boyu Wang
• Jorge Mendez
• Mingbo Cai
• Eric Eaton

We propose a new principle for transfer learning, based on a straightforward intuition: if two domains are similar to each other, the model trained on one domain should also perform well on the other domain, and vice versa. To formalize this intuition, we define the performance gap as a measure of the discrepancy between the source and target domains. We derive generalization bounds for the instance weighting approach to transfer learning, showing that the performance gap can be viewed as an algorithm-dependent regularizer, which controls the model complexity. Our theoretical analysis provides new insight into transfer learning and motivates a set of general, principled rules for designing new instance weighting schemes for transfer learning. These rules lead to gapBoost, a novel and principled boosting approach for transfer learning. Our experimental evaluation on benchmark data sets shows that gapBoost significantly outperforms previous boosting-based transfer learning algorithms.

NeurIPS Conference 2018 Conference Paper

Lifelong Inverse Reinforcement Learning

• Jorge Mendez
• Shashank Shivkumar
• Eric Eaton

Methods for learning from demonstration (LfD) have shown success in acquiring behavior policies by imitating a user. However, even for a single task, LfD may require numerous demonstrations. For versatile agents that must learn many tasks via demonstration, this process would substantially burden the user if each task were learned in isolation. To address this challenge, we introduce the novel problem of lifelong learning from demonstration, which allows the agent to continually build upon knowledge learned from previously demonstrated tasks to accelerate the learning of new tasks, reducing the amount of demonstrations required. As one solution to this problem, we propose the first lifelong learning approach to inverse reinforcement learning, which learns consecutive tasks via demonstration, continually transferring knowledge between tasks to improve performance.

AAAI Conference 2018 Short Paper

Lifelong Learning Networks: Beyond Single Agent Lifelong Learning

• Mohammad Rostami
• Eric Eaton

Lifelong machine learning (LML) is a paradigm to design adaptive agents that can learn in dynamic environments. Current LML algorithms consider a single agent that has centralized access to all data. However, given privacy and security constraints, data might be distributed among multiple agents that can collaborate and learn from collective experience. Our goal is to extend LML from a single agent to a network of multiple agents that collectively learn a series of tasks.

AAMAS Conference 2018 Conference Paper

Modeling Consecutive Task Learning with Task Graph Agendas

• David Isele
• Eric Eaton
• Mark Roberts
• David W. Aha

AAMAS Conference 2018 Conference Paper

Multi-Agent Distributed Lifelong Learning for Collective Knowledge Acquisition

• Mohammad Rostami
• Soheil Kolouri
• Kyungnam Kim
• Eric Eaton

Lifelong machine learning methods acquire knowledge over a series of consecutive tasks, continually building upon their experience. Current lifelong learning algorithms rely upon a single learning agent that has centralized access to all data. In this paper, we extend the idea of lifelong learning from a single agent to a network of multiple agents that collectively learn a series of tasks. Each agent faces some (potentially unique) set of tasks; the key idea is that knowledge learned from these tasks may benefit other agents trying to learn different (but related) tasks. Our Collective Lifelong Learning Algorithm (CoLLA) provides an efficient way for a network of agents to share their learned knowledge in a distributed and decentralized manner, while eliminating the need to share locally observed data. We provide theoretical guarantees for robust performance of the algorithm and empirically demonstrate that CoLLA outperforms existing approaches for distributed multi-task learning on a variety of datasets.

IROS Conference 2016 Conference Paper

Lifelong learning for disturbance rejection on mobile robots

• David Isele
• José-Marcio Luna
• Eric Eaton
• Gabriel Victor de la Cruz
• James Irwin
• Brandon Kallaher
• Matthew E. Taylor

No two robots are exactly the same—even for a given model of robot, different units will require slightly different controllers. Furthermore, because robots change and degrade over time, a controller will need to change over time to remain optimal. This paper leverages lifelong learning in order to learn controllers for different robots. In particular, we show that by learning a set of control policies over robots with different (unknown) motion models, we can quickly adapt to changes in the robot, or learn a controller for a new robot with a unique set of disturbances. Furthermore, the approach is completely model-free, allowing us to apply this method to robots that have not, or cannot, be fully modeled.

IJCAI Conference 2016 Conference Paper

Using Task Features for Zero-Shot Knowledge Transfer in Lifelong Learning

• David Isele
• Mohammad Rostami
• Eric Eaton

Knowledge transfer between tasks can improve the performance of learned models, but requires an accurate estimate of the inter-task relationships to identify the relevant knowledge to transfer. These inter-task relationships are typically estimated based on training data for each task, which is inefficient in lifelong learning settings where the goal is to learn each consecutive task rapidly from as little data as possible. To reduce this burden, we develop a lifelong reinforcement learning method based on coupled dictionary learning that incorporates high-level task descriptors to model the inter-task relationships. We show that using task descriptors improves the performance of the learned task policies, providing both theoretical justification for the benefit and empirical demonstration of the improvement across a variety of dynamical control problems. Given only the descriptor for a new task, the lifelong learner is also able to accurately predict the task policy through zero-shot learning using the coupled dictionary, eliminating the need to pause to gather training data before addressing the task.

IJCAI Conference 2015 Conference Paper

Autonomous Cross-Domain Knowledge Transfer in Lifelong Policy Gradient Reinforcement Learning

• Haitham Bou Ammar
• Eric Eaton
• Jose Marcio Luna
• Paul Ruvolo

Online multi-task learning is an important capability for lifelong learning agents, enabling them to acquire models for diverse tasks over time and rapidly learn new tasks by building upon prior experience. However, recent progress toward lifelong reinforcement learning (RL) has been limited to learning from within a single task domain. For truly versatile lifelong learning, the agent must be able to autonomously transfer knowledge between different task domains. A few methods for cross-domain transfer have been developed, but these methods are computationally inefficient for scenarios where the agent must learn tasks consecutively. In this paper, we develop the first cross-domain lifelong RL framework. Our approach efficiently optimizes a shared repository of transferable knowledge and learns projection matrices that specialize that knowledge to different task domains. We provide rigorous theoretical guarantees on the stability of this approach, and empirically evaluate its performance on diverse dynamical systems. Our results show that the proposed method can learn effectively from interleaved task domains and rapidly acquire high performance in new domains.

ICML Conference 2015 Conference Paper

Safe Policy Search for Lifelong Reinforcement Learning with Sublinear Regret

• Haitham Bou-Ammar
• Rasul Tutunov
• Eric Eaton

Lifelong reinforcement learning provides a promising framework for developing versatile agents that can accumulate knowledge over a lifetime of experience and rapidly learn new tasks by building upon prior knowledge. However, current lifelong learning methods exhibit non-vanishing regret as the amount of experience increases, and include limitations that can lead to suboptimal or unsafe control policies. To address these issues, we develop a lifelong policy gradient learner that operates in an adversarial setting to learn multiple tasks online while enforcing safety constraints on the learned policies. We demonstrate, for the first time, sublinear regret for lifelong policy search, and validate our algorithm on several benchmark dynamical systems and an application to quadrotor control.

AAAI Conference 2015 Conference Paper

Unsupervised Cross-Domain Transfer in Policy Gradient Reinforcement Learning via Manifold Alignment

• Haitham Bou Ammar
• Eric Eaton
• Paul Ruvolo
• Matthew Taylor

The success of applying policy gradient reinforcement learning (RL) to difficult control tasks hinges crucially on the ability to determine a sensible initialization for the policy. Transfer learning methods tackle this problem by reusing knowledge gleaned from solving other related tasks. In the case of multiple task domains, these algorithms require an inter-task mapping to facilitate knowledge transfer across domains. However, there are currently no general methods to learn an inter-task mapping without requiring either background knowledge that is not typically present in RL settings, or an expensive analysis of an exponential number of inter-task mappings in the size of the state and action spaces. This paper introduces an autonomous framework that uses unsupervised manifold alignment to learn intertask mappings and effectively transfer samples between different task domains. Empirical results on diverse dynamical systems, including an application to quadrotor control, demonstrate its effectiveness for cross-domain transfer in the context of policy gradient RL.

AAAI Conference 2014 Conference Paper

Online Multi-Task Gradient Temporal-Difference Learning

• Vishnu Sreenivasan
• Haitham Bou Ammar
• Eric Eaton

ICML Conference 2014 Conference Paper

Online Multi-Task Learning for Policy Gradient Methods

• Haitham Bou-Ammar
• Eric Eaton
• Paul Ruvolo
• Matthew E. Taylor

Policy gradient algorithms have shown considerable recent success in solving high-dimensional sequential decision making tasks, particularly in robotics. However, these methods often require extensive experience in a domain to achieve high performance. To make agents more sample-efficient, we developed a multi-task policy gradient method to learn decision making tasks consecutively, transferring knowledge between tasks to accelerate learning. Our approach provides robust theoretical guarantees, and we show empirically that it dramatically accelerates learning on a variety of dynamical systems, including an application to quadrotor control.

AAAI Conference 2014 Conference Paper

Online Multi-Task Learning via Sparse Dictionary Optimization

• Paul Ruvolo
• Eric Eaton

This paper develops an efficient online algorithm for learning multiple consecutive tasks based on the K- SVD algorithm for sparse dictionary optimization. We first derive a batch multi-task learning method that builds upon K-SVD, and then extend the batch algorithm to train models online in a lifelong learning setting. The resulting method has lower computational complexity than other current lifelong learning algorithms while maintaining nearly identical model performance. Additionally, the proposed method offers an alternate formulation for lifelong learning that supports both task and feature similarity matrices.

AAAI Conference 2013 Conference Paper

Active Task Selection for Lifelong Machine Learning

• Paul Ruvolo
• Eric Eaton

In a lifelong learning framework, an agent acquires knowledge incrementally over consecutive learning tasks, continually building upon its experience. Recent lifelong learning algorithms have achieved nearly identical performance to batch multi-task learning methods while reducing learning time by three orders of magnitude. In this paper, we further improve the scalability of lifelong learning by developing curriculum selection methods that enable an agent to actively select the next task to learn in order to maximize performance on future learning tasks. We demonstrate that active task selection is highly reliable and effective, allowing an agent to learn high performance models using up to 50% fewer tasks than when the agent has no control over the task order. We also explore a variant of transfer learning in the lifelong learning setting in which the agent can focus knowledge acquisition toward a particular target task.

ICML Conference 2013 Conference Paper

ELLA: An Efficient Lifelong Learning Algorithm

• Paul Ruvolo
• Eric Eaton

The problem of learning multiple consecutive tasks, known as lifelong learning, is of great importance to the creation of intelligent, general-purpose, and flexible machines. In this paper, we develop a method for online multi-task learning in the lifelong learning setting. The proposed Efficient Lifelong Learning Algorithm (ELLA) maintains a sparsely shared basis for all task models, transfers knowledge from the basis to learn each new task, and refines the basis over time to maximize performance across all tasks. We show that ELLA has strong connections to both online dictionary learning for sparse coding and state-of-the-art batch multi-task learning methods, and provide robust theoretical performance guarantees. We show empirically that ELLA yields nearly identical performance to batch multi-task learning while learning tasks sequentially in three orders of magnitude (over 1, 000x) less time.

AAAI Conference 2012 Conference Paper

A Spin-Glass Model for Semi-Supervised Community Detection

• Eric Eaton
• Rachael Mansbach

Current modularity-based community detection methods show decreased performance as relational networks become increasingly noisy. These methods also yield a large number of diverse community structures as solutions, which is problematic for applications that impose constraints on the acceptable solutions or in cases where the user is focused on specific communities of interest. To address both of these problems, we develop a semi-supervised spin-glass model that enables current community detection methods to incorporate background knowledge in the forms of individual labels and pairwise constraints. Unlike current methods, our approach shows robust performance in the presence of noise in the relational network, and the ability to guide the discovery process toward specific community structures. We evaluate our algorithm on several benchmark networks and a new political sentiment network representing cooperative events between nations that was mined from news articles over six years.

AAAI Conference 2011 Conference Paper

Selective Transfer Between Learning Tasks Using Task-Based Boosting

• Eric Eaton
• Marie desJardins

The success of transfer learning on a target task is highly dependent on the selected source data. Instance transfer methods reuse data from the source tasks to augment the training data for the target task. If poorly chosen, this source data may inhibit learning, resulting in negative transfer. The current most widely used algorithm for instance transfer, TrAdaBoost, performs poorly when given irrelevant source data. We present a novel task-based boosting technique for instance transfer that selectively chooses the source knowledge to transfer to the target task. Our approach performs boosting at both the instance level and the task level, assigning higher weight to those source tasks that show positive transferability to the target task, and adjusting the weights of individual instances within each source task via AdaBoost. We show that this combination of task- and instance-level boosting significantly improves transfer performance over existing instance transfer algorithms when given a mix of relevant and irrelevant source data, especially for small amounts of data on the target task.

AAAI Conference 2010 Conference Paper

Interactive Learning Using Manifold Geometry

• Eric Eaton
• Gary Holness
• Daniel McFarlane

We present an interactive learning method that enables a user to iteratively refine a regression model. The user examines the output of the model, visualized as the vertical axis of a 2D scatterplot, and provides corrections by repositioning individual data instances to the correct output level. Each repositioned data instance acts as a control point for altering the learned model, using the geometry underlying the data. We capture the underlying structure of the data as a manifold, on which we compute a set of basis functions as the foundation for learning. Our results show that manifold-based interactive learning improves performance monotonically with each correction, outperforming alternative approaches.

ICML Conference 2006 Conference Paper

Learning user preferences for sets of objects

• Marie desJardins
• Eric Eaton
• Kiri Wagstaff

AAAI Conference 2006 Short Paper

Multi-Resolution Learning for Knowledge Transfer

• Eric Eaton

Related objects may look similar at low-resolutions; differences begin to emerge naturally as the resolution is increased. By learning across multiple resolutions of input, knowledge can be transfered between related objects. My dissertation develops this idea and applies it to the problem of multitask transfer learning.