Glen Berseth

Pablo Samuel Castro

Aaron Courville

Plasticity, or the ability of an agent to adapt to new tasks, environments, or distributions, is crucial for continual learning. In this pap… (voir plus)er, we study the loss of plasticity in deep continual RL from the lens of churn: network output variability induced by the data in each training batch. We demonstrate that (1) the loss of plasticity is accompanied by the exacerbation of churn due to the gradual rank decrease of the Neural Tangent Kernel (NTK) matrix; (2) reducing churn helps prevent rank collapse and adjusts the step size of regular RL gradients adaptively. Moreover, we introduce Continual Churn Approximated Reduction (C-CHAIN) and demonstrate it improves learning performance and outperforms baselines in a diverse range of continual learning environments on OpenAI Gym Control, ProcGen, DeepMind Control Suite, and MinAtar benchmarks.

2025-10-06

Proceedings of the 42nd International Conference on Machine Learning (publié)

proceedings.mlr.press

Outsourced Diffusion Sampling: Efficient Posterior Inference in Latent Spaces of Generative Models

Any well-behaved generative model over a variable …

2025-10-06

Proceedings of the 42nd International Conference on Machine Learning (publié)

proceedings.mlr.press

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models

Yoshua Bengio

Brian R. Bartoldson

Bhavya Kailkhura

Guillaume Lajoie

Nikolay Malkin

Moksh J. Jain

Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference… (voir plus) to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines a population of candidate reasoning chains through aggregation of subsets to yield a population of improved solutions, which are then used as the candidate pool for the next iteration. RSA exploits the rich information embedded in the reasoning chains -- not just the final answers -- and enables bootstrapping from partially correct intermediate steps within different chains of thought. Empirically, RSA delivers substantial performance gains with increasing compute budgets across diverse tasks, model families and sizes. Notably, RSA enables Qwen3-4B-Instruct-2507 to achieve competitive performance with larger reasoning models, including DeepSeek-R1 and o3-mini (high), while outperforming purely parallel and sequential scaling strategies across AIME-25, HMMT-25, Reasoning Gym, LiveCodeBench-v6, and SuperGPQA. We further demonstrate that training the model to combine solutions via a novel aggregation-aware reinforcement learning approach yields significant performance gains. Code available at https://github.com/HyperPotatoNeo/RSA.

2025-09-30

ArXiv (prépublication)

arxiv.org

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models

Yoshua Bengio

Brian R. Bartoldson

Bhavya Kailkhura

Guillaume Lajoie

Nikolay Malkin

Moksh J. Jain

Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference… (voir plus) to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines a population of candidate reasoning chains through aggregation of subsets to yield a population of improved solutions, which are then used as the candidate pool for the next iteration. RSA exploits the rich information embedded in the reasoning chains -- not just the final answers -- and enables bootstrapping from partially correct intermediate steps within different chains of thought. Empirically, RSA delivers substantial performance gains with increasing compute budgets across diverse tasks, model families and sizes. Notably, RSA enables Qwen3-4B-Instruct-2507 to achieve competitive performance with larger reasoning models, including DeepSeek-R1 and o3-mini (high), while outperforming purely parallel and sequential scaling strategies across AIME-25, HMMT-25, Reasoning Gym, LiveCodeBench-v6, and SuperGPQA. We further demonstrate that training the model to combine solutions via a novel aggregation-aware reinforcement learning approach yields significant performance gains. Code available at https://github.com/HyperPotatoNeo/RSA.

2025-09-30

ArXiv (prépublication)

arxiv.org

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models

Yoshua Bengio

Brian R. Bartoldson

Bhavya Kailkhura

Guillaume Lajoie

Nikolay Malkin

Moksh J. Jain

Test-time scaling methods improve the capabilities of large language models (LLMs) by increasing the amount of compute used during inference… (voir plus) to make a prediction. Inference-time compute can be scaled in parallel by choosing among multiple independent solutions or sequentially through self-refinement. We propose Recursive Self-Aggregation (RSA), a test-time scaling method inspired by evolutionary methods that combines the benefits of both parallel and sequential scaling. Each step of RSA refines a population of candidate reasoning chains through aggregation of subsets to yield a population of improved solutions, which are then used as the candidate pool for the next iteration. RSA exploits the rich information embedded in the reasoning chains -- not just the final answers -- and enables bootstrapping from partially correct intermediate steps within different chains of thought. Empirically, RSA delivers substantial performance gains with increasing compute budgets across diverse tasks, model families and sizes. Notably, RSA enables Qwen3-4B-Instruct-2507 to achieve competitive performance with larger reasoning models, including DeepSeek-R1 and o3-mini (high), while outperforming purely parallel and sequential scaling strategies across AIME-25, HMMT-25, Reasoning Gym, LiveCodeBench-v6, and SuperGPQA. We further demonstrate that training the model to combine solutions via a novel aggregation-aware reinforcement learning approach yields significant performance gains. Code available at https://github.com/HyperPotatoNeo/RSA.

2025-09-30

ArXiv (prépublication)

arxiv.org

Task Robustness via Re-Labelling Vision-Action Robot Data

Artur Kuramshin

Özgür Aslan

Cyrus Neary

The recent trend in scaling models for robot learning has resulted in impressive policies that can perform various manipulation tasks and ge… (voir plus)neralize to novel scenarios. However, these policies continue to struggle with following instructions, likely due to the limited linguistic and action sequence diversity in existing robotics datasets. This paper introduces

2025-09-06

robot-learning.org/CoRL/2025/Workshop/Robot_Data (publié)

RoboArena: Distributed Real-World Evaluation of Generalist Robot Policies

Pranav Atreya

Karl Pertsch

Tony Lee

Moo Jin Kim

Arhan Jain

Artur Kuramshin

Cyrus Neary

Edward S. Hu

Kanav Arora

Kirsty Ellis

Luca Macesanu

Matthew Leonard

Meedeum Cho

Özgür Aslan

Shivin Dass

Tony Wang

Xingfang Yuan

Abhishek Gupta

Dinesh Jayaraman

Glen Berseth … (voir 6 de plus)

Kostas Daniilidis

Roberto Martín-Martín

Youngwoon Lee

Percy Liang

Chelsea Finn

Sergey Levine

2025-08-08

robot-learning.org/CoRL/2025/Conference (présentation orale)

Adaptive Resolution Residual Networks — Generalizing Across Resolutions Easily and Efficiently

Léa Demeule

Mahtab Sandhu

The majority of signal data captured in the real world uses numerous sensors with different resolutions. In practice, most deep learning arc… (voir plus)hitectures are fixed-resolution; they consider a single resolution at training and inference time. This is convenient to implement but fails to fully take advantage of the diverse signal data that exists. In contrast, other deep learning architectures are adaptive-resolution; they directly allow various resolutions to be processed at training and inference time. This provides computational adaptivity but either sacrifices robustness or compatibility with mainstream layers, which hinders their use. In this work, we introduce Adaptive Resolution Residual Networks (ARRNs) to surpass this tradeoff. We construct ARRNs from Laplacian residuals, which serve as generic adaptive-resolution adapters for fixed-resolution layers. We use smoothing filters within Laplacian residuals to linearly separate input signals over a series of resolution steps. We can thereby skip Laplacian residuals to cast high-resolution ARRNs into low-resolution ARRNs that are computationally cheaper yet numerically identical over low-resolution signals. We guarantee this result when Laplacian residuals are implemented with perfect smoothing kernels. We complement this novel component with Laplacian dropout, which randomly omits Laplacian residuals during training. This regularizes for robustness to a distribution of lower resolutions. This also regularizes for numerical errors that may occur when Laplacian residuals are implemented with approximate smoothing kernels. We provide a solid grounding for the advantageous properties of ARRNs through a theoretical analysis based on neural operators, and empirically show that ARRNs embrace the challenge posed by diverse resolutions with computational adaptivity, robustness, and compatibility with mainstream layers.

2025-07-17

TMLR (accepté)

Curiosity-Driven Exploration via Temporal Contrastive Learning

Faisal Mohamed

Catherine Ji

Benjamin Eysenbach

Effective exploration in reinforcement learning requires keeping track not just of where the agent has been, but also of how the agent think… (voir plus)s about and represents the world: an agent should explore states that enable it to learn powerful representations. Temporal representations can include the information required to solve any potential task while avoiding the computational cost of reconstruction. In this paper, we propose an exploration method that uses temporal contrastive representations to drive exploration, maximizing coverage as seen through the lens of these temporal representations. We demonstrate complex exploration behaviors in locomotion, manipulation, and embodied-AI tasks, revealing previously unknown capabilities and behaviors once achievable only via extrinsic rewards.

2025-07-01

rl-conference.cc/RLC/2025/Workshop/RLBrew (publié)

Curiosity-Driven Exploration via Temporal \\ Contrastive Learning

Faisal Mohamed

Catherine Ji

Benjamin Eysenbach

Exploration remains a key challenge in reinforcement learning (RL), especially in long-horizon tasks and environments with high-dimensional … (voir plus)observations. A common strategy for effective exploration is to promote state coverage or novelty, which often involves estimating the agent's state visitation distribution. In this paper, we propose \textbf{C}uriosity-Driven Exploration via \textbf{Te}mporal \textbf{C}ontrastive Learning (\methodName), an exploration method based on temporal contrastive learning that rewards agents for reaching states with unexpected futures. This incentivizes uncovering meaningful less-visited states. \methodName is simple and does not require explicit density or uncertainty estimation, while learning representations aligned with the RL objective. It consistently outperforms standard baselines in complex mazes using different embodiments (Ant and Humanoid) and robotic manipulation tasks, while also yielding more diverse behaviors in Craftax without requiring task-specific information.

2025-07-01

rl-conference.cc/RLC/2025/Workshop/RLBrew (publié)

Is Exploration or Optimization the Problem for Deep Reinforcement Learning?

2025-07-01

rl-conference.cc/RLC/2025/Workshop/Finding_the_Frame (publié)