PPO（Proximal Policy Optimization） 这个后来在 RLHF 和大模型训练中被广泛使用的经典算法，当年曾被 NIPS 2017 拒之门外。 这件事最近由 PPO 作者 John Schulman 本人提起。他只用一句话概括了这段往事：PPO，曾经被 NIPS 2017 拒了。

As cities continue to expand, railways are expected to become an important component of urban mobility systems. Compared with ...

COLREGs-compliant ship collision avoidance strategy based on proximal policy optimization algorithm - Frontiers in Marine Science: The safe and efficient collision avoidance of multiple ships is ...

Abstract: The challenge of navigating unmanned aerial vehicles (UAVs) can be effectively tackled through the application of reinforcement learning (RL) methodologies. Nonetheless, the baseline ...

Reinforcement learning (RL) has witnessed tremendous advances in recent years, enabling agents to master tasks ranging from video games to robotics. However, designing stable, sample-efficient ...

In recent years, with the continuous development of reinforcement learning (RL), we have seen promising results in processing continuous action RL tasks 1,2,3,4,5. In dealing with some continuous ...

Transformer-based models have revolutionized natural language processing. Models like GPT-4, BERT, and T5 dominate NLP applications in 2024, powering language translation, text summarization, and ...

The rise of artificial intelligence (AI) deep learning algorithms is helping to accelerate brain-computer interfaces (BCIs). Published in this month’s Nature Neuroscience is new research that shows ...

The performance of the state-of-the-art Deep Reinforcement algorithms such as Proximal Policy Optimization, Twin Delayed Deep Deterministic Policy Gradient, and Soft Actor-Critic for generating a ...