paper-with-me

홈 › Papers

Neural Auto-Curricula

2021-06-04 · Xidong Feng, Oliver Slumbers, Ziyu Wan, Bo Liu, Stephen Mcaleer, Ying Wen, Jun Wang, Yaodong Yang

When solving two-player zero-sum games, multi-agent reinforcement learning (MARL) algorithms often create populations of agents where, at each iteration, a new agent is discovered as the best response to a mixture over the opponent population. Within such a process, the update rules of "who to compete with" (i.e., the opponent mixture) and "how to beat them" (i.e., finding best responses) are underpinned by manually developed game theoretical principles such as fictitious play and Double Oracle. In this paper, we introduce a novel framework -- Neural Auto-Curricula (NAC) -- that leverages meta-gradient descent to automate the discovery of the learning update rule without explicit human design. Specifically, we parameterise the opponent selection module by neural networks and the best-response module by optimisation subroutines, and update their parameters solely via interaction with the game engine, where both players aim to minimise their exploitability. Surprisingly, even without human design, the discovered MARL algorithms achieve competitive or even better performance with the state-of-the-art population-based game solvers (e.g., PSRO) on Games of Skill, differentiable Lotto, non-transitive Mixture Games, Iterated Matching Pennies, and Kuhn Poker. Additionally, we show that NAC is able to generalise from small games to large games, for example training on Kuhn Poker and outperforming PSRO on Leduc Poker. Our work inspires a promising future direction to discover general MARL algorithms solely from data.

📄 PDF Abstract BibTeX arXiv:2106.02745

Code (1)

waterhorse1/nac 공식 구현 pytorch

Tasks

Multi-agent Reinforcement Learning

Similar Papers 제목 키워드 기반

Automated curricula through setter-solver interactions

2019-09-27 · Sebastien Racaniere, Andrew K. Lampinen, Adam Santoro, David P. Reichert 외

Reinforcement learning algorithms use correlations between policies and rewards to improve agent performance. But in dynamic or sparsely rewarding environments these correlations are often too small, or rewarding events …

Reinforcement Learning

HiCu: Leveraging Hierarchy for Curriculum Learning in Automated ICD Coding

2022-08-03 · Weiming Ren, Ruijing Zeng, Tongzi Wu, Tianshu Zhu 외

There are several opportunities for automation in healthcare that can improve clinician throughput. One such example is assistive tools to document diagnosis codes when clinicians write notes. We study the automation of …

Medical Code PredictionMulti-Label ClassificationMUlTI-LABEL-ClASSIFICATION

Automatic Curricula via Expert Demonstrations

2021-06-16 · Siyu Dai, Andreas Hofmann, Brian Williams

We propose Automatic Curricula via Expert Demonstrations (ACED), a reinforcement learning (RL) approach that combines the ideas of imitation learning and curriculum learning in order to solve challenging robotic manipula…

Imitation LearningReinforcement Learning (RL)

Automated curriculum generation through setter-solver interactions

2020-05-01 · ICLR 2020 1 · Sebastien Racaniere, Andrew Lampinen, Adam Santoro, David Reichert 외

Reinforcement learning algorithms use correlations between policies and rewards to improve agent performance. But in dynamic or sparsely rewarding environments these correlations are often too small, or rewarding even…

A Probabilistic Interpretation of Self-Paced Learning with Applications to Reinforcement Learning

2021-02-25 · Pascal Klink, Hany Abdulsamad, Boris Belousov, Carlo D'Eramo 외

Across machine learning, the use of curricula has shown strong empirical potential to improve learning from data by avoiding local optima of training objectives. For reinforcement learning (RL), curricula are especially …

reinforcement-learningReinforcement Learning (RL)