All Language Models Large and Small

no code implementations • 19 Feb 2024 • Zhixun Chen, Yali Du, David Mguni

Theoretically, we prove LONDI learns the subset of system states to activate the LLM required to solve the task.

Decision Making

ChessGPT: Bridging Policy Learning and Language Modeling

1 code implementation • NeurIPS 2023 • Xidong Feng, Yicheng Luo, Ziyan Wang, Hongrui Tang, Mengyue Yang, Kun Shao, David Mguni, Yali Du, Jun Wang

Thus, we propose ChessGPT, a GPT model bridging policy learning and language modeling by integrating data from these two sources in Chess games.

Decision Making Language Modelling

Ensemble Value Functions for Efficient Exploration in Multi-Agent Reinforcement Learning

no code implementations • 7 Feb 2023 • Lukas Schäfer, Oliver Slumbers, Stephen Mcaleer, Yali Du, Stefano V. Albrecht, David Mguni

In this work, we propose ensemble value functions for multi-agent exploration (EMAX), a general framework to seamlessly extend value-based MARL algorithms with ensembles of value functions.

Efficient Exploration Multi-agent Reinforcement Learning +2

Taming Multi-Agent Reinforcement Learning with Estimator Variance Reduction

no code implementations • 2 Sep 2022 • Taher Jafferjee, Juliusz Ziomek, Tianpei Yang, Zipeng Dai, Jianhong Wang, Matthew Taylor, Kun Shao, Jun Wang, David Mguni

Centralised training with decentralised execution (CT-DE) serves as the foundation of many leading multi-agent reinforcement learning (MARL) algorithms.

Multi-agent Reinforcement Learning reinforcement-learning +3

Timing is Everything: Learning to Act Selectively with Costly Actions and Budgetary Constraints

no code implementations • 31 May 2022 • David Mguni, Aivar Sootla, Juliusz Ziomek, Oliver Slumbers, Zipeng Dai, Kun Shao, Jun Wang

In this paper, we introduce a reinforcement learning (RL) framework named \textbf{L}earnable \textbf{I}mpulse \textbf{C}ontrol \textbf{R}einforcement \textbf{A}lgorithm (LICRA), for learning to optimally select both when to act and which actions to take when actions incur costs.

Reinforcement Learning (RL)

SEREN: Knowing When to Explore and When to Exploit

no code implementations • 30 May 2022 • Changmin Yu, David Mguni, Dong Li, Aivar Sootla, Jun Wang, Neil Burgess

Efficient reinforcement learning (RL) involves a trade-off between "exploitative" actions that maximise expected reward and "explorative'" ones that sample unvisited states.

Reinforcement Learning (RL)

On the Convergence of Fictitious Play: A Decomposition Approach

no code implementations • 3 May 2022 • Yurong Chen, Xiaotie Deng, Chenchen Li, David Mguni, Jun Wang, Xiang Yan, Yaodong Yang

Fictitious play (FP) is one of the most fundamental game-theoretical learning frameworks for computing Nash equilibrium in $n$-player games, which builds the foundation for modern multi-agent learning algorithms.

Saute RL: Almost Surely Safe Reinforcement Learning Using State Augmentation

1 code implementation • 14 Feb 2022 • Aivar Sootla, Alexander I. Cowen-Rivers, Taher Jafferjee, Ziyan Wang, David Mguni, Jun Wang, Haitham Bou-Ammar

Satisfying safety constraints almost surely (or with probability one) can be critical for the deployment of Reinforcement Learning (RL) in real-life applications.

reinforcement-learning Reinforcement Learning (RL) +1

DESTA: A Framework for Safe Reinforcement Learning with Markov Games of Intervention

no code implementations • 27 Oct 2021 • David Mguni, Usman Islam, Yaqi Sun, Xiuling Zhang, Joel Jennings, Aivar Sootla, Changmin Yu, Ziyan Wang, Jun Wang, Yaodong Yang

In this paper, we introduce a new generation of RL solvers that learn to minimise safety violations while maximising the task reward to the extent that can be tolerated by the safe policy.

OpenAI Gym reinforcement-learning +3

On the Complexity of Computing Markov Perfect Equilibrium in General-Sum Stochastic Games

no code implementations • 4 Sep 2021 • Xiaotie Deng, Ningyuan Li, David Mguni, Jun Wang, Yaodong Yang

Similar to the role of Markov decision processes in reinforcement learning, Stochastic Games (SGs) lay the foundation for the study of multi-agent reinforcement learning (MARL) and sequential agent interactions.

Multi-agent Reinforcement Learning reinforcement-learning +1

Learning to Shape Rewards using a Game of Two Partners

no code implementations • 16 Mar 2021 • David Mguni, Taher Jafferjee, Jianhong Wang, Nicolas Perez-Nieves, Tianpei Yang, Matthew Taylor, Wenbin Song, Feifei Tong, Hui Chen, Jiangcheng Zhu, Jun Wang, Yaodong Yang

Reward shaping (RS) is a powerful method in reinforcement learning (RL) for overcoming the problem of sparse or uninformative rewards.

reinforcement-learning Reinforcement Learning (RL) +1

Multi-Agent Determinantal Q-Learning

1 code implementation • ICML 2020 • Yaodong Yang, Ying Wen, Li-Heng Chen, Jun Wang, Kun Shao, David Mguni, Wei-Nan Zhang

Though practical, current methods rely on restrictive assumptions to decompose the centralized value function across agents for execution.

Q-Learning