Foundation models are first pre-trained on vast unsupervised datasets and then fine-tuned
on labeled data. Reinforcement learning, notably from human feedback (RLHF), can further …

Despite recent progress in reinforcement learning (RL) from raw pixel data, sample
inefficiency continues to present a substantial obstacle. Prior works have attempted to …

Despite the remarkable progress made in visual reinforcement learning (RL) in recent years,
sample inefficiency remains a major challenge. Many existing approaches attempt to …

Nowadays, deep reinforcement learning (DRL) has made remarkable achievements in the
research of unmanned aerial vehicle (UAV) applications. However, much of the current …

为了提升具有高维动作空间的复杂连续控制任务的性能和样本效率, 提出一种基于Transformer
的状态− 动作− 奖赏预测表征学习框架(Transformer-based state-action-reward prediction …

The ability to learn robust policies while generalizing over large discrete action spaces is an
open challenge for intelligent systems, especially in noisy environments that face the curse …

Machine autonomy in automatic control often relies on reinforcement learning (RL) for robot
control. However, RL agents struggle with complex long-sequence tasks due to catastrophic …