Scott Fujimoto

Scalable Option Learning in High-Throughput Environments

Mikael Henaff

Michael Matthews

Michael G. Rabbat

Hierarchical reinforcement learning (RL) has the potential to enable effective decision-making over long timescales. Existing approaches, wh… (voir plus)ile promising, have yet to realize the benefits of large-scale training. In this work, we identify and solve several key challenges in scaling hierarchical RL to high-throughput environments. We propose Scalable Option Learning (SOL), a highly scalable hierarchical RL algorithm which achieves a 25x higher throughput compared to existing hierarchical methods. We train our hierarchical agents using 20 billion frames of experience on the complex game of NetHack, significantly surpassing flat agents and demonstrating positive scaling trends. We also validate our algorithm on MiniHack and Mujoco environments, showcasing its general applicability. Our code is open sourced at github.com/facebookresearch/sol.

2025-08-29

ArXiv (prépublication)

Sparse-Reg: Improving Sample Complexity in Offline Reinforcement Learning using Sparsity

Samin Yeasar Arnob

In this paper, we investigate the use of small datasets in the context of offline reinforcement learning (RL). While many common offline RL … (voir plus)benchmarks employ datasets with over a million data points, many offline RL applications rely on considerably smaller datasets. We show that offline RL algorithms can overfit on small datasets, resulting in poor performance. To address this challenge, we introduce"Sparse-Reg": a regularization technique based on sparsity to mitigate overfitting in offline reinforcement learning, enabling effective learning in limited data settings and outperforming state-of-the-art baselines in continuous control.

2025-05-31

arXiv (publié)

Generalizable Imitation Learning Through Pre-Trained Representations

Wei-Di Chang

Francois Hogan

Gregory Dudek

In this paper we leverage self-supervised vision transformer models and their emergent semantic abilities to improve the generalization abil… (voir plus)ities of imitation learning policies. We introduce BC-ViT, an imitation learning algorithm that leverages rich DINO pre-trained Visual Transformer (ViT) patch-level embeddings to obtain better generalization when learning through demonstrations. Our learner sees the world by clustering appearance features into semantic concepts, forming stable keypoints that generalize across a wide range of appearance variations and object types. We show that this representation enables generalized behaviour by evaluating imitation learning across a diverse dataset of object manipulation tasks. Our method, data and evaluation approach are made available to facilitate further study of generalization in Imitation Learners.

2025-05-18

2025 IEEE International Conference on Robotics and Automation (ICRA) (publié)

Towards General-Purpose Model-Free Reinforcement Learning

Pierluca D'Oro

Amy Zhang

Yuandong Tian

Michael G. Rabbat

Reinforcement learning (RL) promises a framework for near-universal problem-solving. In practice however, RL algorithms are often tailored t… (voir plus)o specific benchmarks, relying on carefully tuned hyperparameters and algorithmic choices. Recently, powerful model-based RL methods have shown impressive general results across benchmarks but come at the cost of increased complexity and slow run times, limiting their broader applicability. In this paper, we attempt to find a unifying model-free deep RL algorithm that can address a diverse class of domains and problem settings. To achieve this, we leverage model-based representations that approximately linearize the value function, taking advantage of the denser task objectives used by model-based RL while avoiding the costs associated with planning or simulated trajectories. We evaluate our algorithm, MR.Q, on a variety of common RL benchmarks with a single set of hyperparameters and show a competitive performance against domain-specific and general baselines, providing a concrete step towards building general-purpose model-free deep RL algorithms.

2025-01-21

ICLR.cc/2025/Conference (spotlight)

openreview.net

Fairness in Reinforcement Learning with Bisimulation Metrics

Sahand Rezaei-Shoshtari

Ensuring long-term fairness is crucial when developing automated decision making systems, specifically in dynamic and sequential environment… (voir plus)s. By maximizing their reward without consideration of fairness, AI agents can introduce disparities in their treatment of groups or individuals. In this paper, we establish the connection between bisimulation metrics and group fairness in reinforcement learning. We propose a novel approach that leverages bisimulation metrics to learn reward functions and observation dynamics, ensuring that learners treat groups fairly while reflecting the original problem. We demonstrate the effectiveness of our method in addressing disparities in sequential decision making problems through empirical evaluation on a standard fairness benchmark consisting of lending and college admission scenarios.

2024-12-21

ArXiv (prépublication)

Imitation Learning from Observation through Optimal Transport

Wei-Di Chang

Gregory Dudek

2024-05-13

rl-conference.cc/RLC/2024/Conference (publié)

openreview.net

For SALE: State-Action Representation Learning for Deep Reinforcement Learning

Wei-Di Chang

Edward J. Smith

Shixiang Shane Gu

In the field of reinforcement learning (RL), representation learning is a proven tool for complex image-based tasks, but is often overlooked… (voir plus) for environments with low-level states, such as physical control problems. This paper introduces SALE, a novel approach for learning embeddings that model the nuanced interaction between state and action, enabling effective representation learning from low-level states. We extensively study the design space of these embeddings and highlight important design considerations. We integrate SALE and an adaptation of checkpoints for RL into TD3 to form the TD7 algorithm, which significantly outperforms existing continuous control algorithms. On OpenAI gym benchmark tasks, TD7 has an average performance gain of 276.7% and 50.7% over TD3 at 300k and 5M time steps, respectively, and works in both the online and offline settings.

2023-09-20

NeurIPS.cc/2023/Conference (poster)

openreview.net

Why Should I Trust You, Bellman? The Bellman Error is a Poor Replacement for Value Error

Ofir Nachum

Shixiang Shane Gu

In this work, we study the use of the Bellman equation as a surrogate objective for value prediction accuracy. While the Bellman equation is… (voir plus) uniquely solved by the true value function over all state-action pairs, we find that the Bellman error (the difference between both sides of the equation) is a poor proxy for the accuracy of the value function. In particular, we show that (1) due to cancellations from both sides of the Bellman equation, the magnitude of the Bellman error is only weakly related to the distance to the true value function, even when considering all state-action pairs, and (2) in the finite data regime, the Bellman equation can be satisfied exactly by infinitely many suboptimal solutions. This means that the Bellman error can be minimized without improving the accuracy of the value function. We demonstrate these phenomena through a series of propositions, illustrative toy examples, and empirical analysis in standard benchmark domains.

2022-06-27

Proceedings of the 39th International Conference on Machine Learning (publié)

proceedings.mlr.press

IL-flOw: Imitation Learning from Observation using Normalizing Flows

Wei-Di Chang

Juan Higuera

Gregory Dudek

2022-05-18

ArXiv (prépublication)

A Deep Reinforcement Learning Approach to Marginalized Importance Sampling with the Successor Representation

Marginalized importance sampling (MIS), which measures the density ratio between the state-action occupancy of a target policy and that of a… (voir plus) sampling distribution, is a promising approach for off-policy evaluation. However, current state-of-the-art MIS methods rely on complex optimization tricks and succeed mostly on simple toy problems. We bridge the gap between MIS and deep reinforcement learning by observing that the density ratio can be computed from the successor representation of the target policy. The successor representation can be trained through deep reinforcement learning methodology and decouples the reward optimization from the dynamics of the environment, making the resulting algorithm stable and applicable to high-dimensional domains. We evaluate the empirical performance of our approach on a variety of challenging Atari and MuJoCo environments.

2021-06-30

Proceedings of the 38th International Conference on Machine Learning (publié)

proceedings.mlr.press

An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay

Prioritized Experience Replay (PER) is a deep reinforcement learning technique in which agents learn from transitions sampled with non-unifo… (voir plus)rm probability proportionate to their temporal-difference error. We show that any loss function evaluated with non-uniformly sampled data can be transformed into another uniformly sampled loss function with the same expected gradient. Surprisingly, we find in some environments PER can be replaced entirely by this new loss function without impact to empirical performance. Furthermore, this relationship suggests a new branch of improvements to PER by correcting its uniformly sampled loss function equivalent. We demonstrate the effectiveness of our proposed modifications to PER and the equivalent loss function in several MuJoCo and Atari environments.

2019-12-31

Advances in Neural Information Processing Systems 33 (NeurIPS 2020) (publié)