Reinforcement Learning Progress
Editorial Pick: OpenAI 使用 PPO 算法训练 Dota 智能体取得突破，展示了深度强化学习在复杂决策问题上的潜力。
Research Papers
All TopicsNotable AI papers and academic breakthroughs: architecture innovations, empirical measurements, and theoretical advances.
281 featured itemsRelated TopicsReasoning & LogicSafety & AlignmentData & Training
Latest Featured
Items 281–281 of 2816月26日星期二周二