H Yuan, Z Mu, F Xie, Z Lu - The Twelfth International Conference on …, 2024 - openreview.net
Pre-training on task-agnostic large datasets is a promising approach for enhancing the sample efficiency of reinforcement learning (RL) in solving complex tasks. We present …
Decision-making is a dynamic process requiring perception, memory, and reasoning to make choices and find optimal policies. Traditional approaches to decision-making suffer …
Q Lv, X Deng, G Chen, MY Wang, L Nie - arXiv preprint arXiv:2406.05427, 2024 - arxiv.org
While the conditional sequence modeling with the transformer architecture has demonstrated its effectiveness in dealing with offline reinforcement learning (RL) tasks, it is …
H Wang, Y Pan, F Sun, S Liu, K Talluri, G Chen… - arXiv preprint arXiv …, 2024 - arxiv.org
In this paper, we consider the supervised pretrained transformer for a class of sequential decision-making problems. The class of considered problems is a subset of the general …
Decision Transformers (DTs) have been highly effective for offline reinforcement learning (RL) tasks, successfully modeling the sequences of actions in a given set of demonstrations …
Despite the promising performance of Decision Transformers (DT) on a wide range of tasks, recent studies have found that the performance of DT may largely be dependent on the …
Z Mu, X Xiao - Peking University Course: Cognitive Reasoning - openreview.net
The Theory of Mind (ToM) ability in multi-agent systems is crucial for coordinating cooperation and understanding communication. ToM involves the capacity to reason about …
Sequence models have emerged as an alternate paradigm for offline Reinforcement Learning (RL) with their remarkable generative capabilities. However, it struggles in cases …