paper-with-me

Papers

Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control

2024-05-25 · Michal Nauman, Mateusz Ostaszewski, Krzysztof Jankowski, Piotr Miłoś, Marek Cygan

Sample efficiency in Reinforcement Learning (RL) has traditionally been driven by algorithmic enhancements. In this work, we demonstrate that scaling can also lead to substantial improvements. We conduct a thorough investigation into the interplay of scaling model capacity and domain-specific RL enhancements. These empirical findings inform the design choices underlying our proposed BRO (Bigger, Regularized, Optimistic) algorithm. The key innovation behind BRO is that strong regularization allows for effective scaling of the critic networks, which, paired with optimistic exploration, leads to superior performance. BRO achieves state-of-the-art results, significantly outperforming the leading model-based and model-free algorithms across 40 complex tasks from the DeepMind Control, MetaWorld, and MyoSuite benchmarks. BRO is the first model-free algorithm to achieve near-optimal policies in the notoriously challenging Dog and Humanoid tasks.

📄 PDF Abstract BibTeX arXiv:2405.16158

Code (1)

naumix/BiggerRegularizedOptimistic 공식 구현 jax

Tasks

continuous-controlContinuous ControlReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Bigger, Better, Faster: Human-level Atari with human-level efficiency

2023-05-30 · Max Schwarzer, Johan Obando-Ceron, Aaron Courville, Marc Bellemare 외

We introduce a value-based RL agent, which we call BBF, that achieves super-human performance in the Atari 100K benchmark. BBF relies on scaling the neural networks used for value estimation, as well as a number of other…

Atari Games 100k

Slamming: Training a Speech Language Model on One GPU in a Day

2025-02-19 · Gallil Maimon, Avishai Elmakies, Yossi Adi

We introduce Slam, a recipe for training high-quality Speech Language Models (SLMs) on a single academic GPU in 24 hours. We do so through empirical analysis of model initialisation and architecture, synthetic training d…

GPULanguage ModelingLanguage Modelling

Optimism Without Regularization: Constant Regret in Zero-Sum Games

2025-06-20 · John Lazarsfeld, Georgios Piliouras, Ryann Sim, Stratis Skoulakis

This paper studies the optimistic variant of Fictitious Play for learning in two-player zero-sum games. While it is known that Optimistic FTRL -- a regularized algorithm with a bounded stepsize parameter -- obtains const…

Optimistically Optimistic Exploration for Provably Efficient Infinite-Horizon Reinforcement and Imitation Learning

2025-02-19 · Antoine Moulin, Gergely Neu, Luca Viano

We study the problem of reinforcement learning in infinite-horizon discounted linear Markov decision processes (MDPs), and propose the first computationally efficient algorithm achieving near-optimal regret guarantees in…

Imitation Learning

Bigger is not Always Better: Scaling Properties of Latent Diffusion Models

2024-04-01 · Kangfu Mei, Zhengzhong Tu, Mauricio Delbracio, Hossein Talebi 외

We study the scaling properties of latent diffusion models (LDMs) with an emphasis on their sampling efficiency. While improved network architecture and inference algorithms have shown to effectively boost sampling effic…