Picture for Jakub Grudzien Kuba

Jakub Grudzien Kuba

IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies

Apr 20, 2023
Philippe Hansen-Estruch, Ilya Kostrikov, Michael Janner, Jakub Grudzien Kuba, Sergey Levine

Figure 1 for IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies

Figure 2 for IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies

Figure 3 for IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies

Figure 4 for IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies

* 11 Pages, 6 Figures, 3 Tables

Access Paper or Ask Questions

Heterogeneous-Agent Reinforcement Learning

Apr 19, 2023
Yifan Zhong, Jakub Grudzien Kuba, Siyi Hu, Jiaming Ji, Yaodong Yang

Figure 1 for Heterogeneous-Agent Reinforcement Learning

Figure 2 for Heterogeneous-Agent Reinforcement Learning

Figure 3 for Heterogeneous-Agent Reinforcement Learning

Figure 4 for Heterogeneous-Agent Reinforcement Learning

* arXiv admin note: text overlap with arXiv:2208.01682

Access Paper or Ask Questions

Discovered Policy Optimisation

Oct 13, 2022
Chris Lu, Jakub Grudzien Kuba, Alistair Letcher, Luke Metz, Christian Schroeder de Witt, Jakob Foerster

Figure 1 for Discovered Policy Optimisation

Figure 2 for Discovered Policy Optimisation

Figure 3 for Discovered Policy Optimisation

Figure 4 for Discovered Policy Optimisation

* NeurIPS 2022

Access Paper or Ask Questions

Understanding Value Decomposition Algorithms in Deep Cooperative Multi-Agent Reinforcement Learning

Feb 16, 2022
Zehao Dou, Jakub Grudzien Kuba, Yaodong Yang

* 37 pages

Access Paper or Ask Questions

Mirror Learning: A Unifying Framework of Policy Optimisation

Feb 02, 2022
Jakub Grudzien Kuba, Christian Schroeder de Witt, Jakob Foerster

Figure 1 for Mirror Learning: A Unifying Framework of Policy Optimisation

Figure 2 for Mirror Learning: A Unifying Framework of Policy Optimisation

Figure 3 for Mirror Learning: A Unifying Framework of Policy Optimisation

Access Paper or Ask Questions