The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret Lukas Fluri,Leon Lang,Alessandro Abate,Patrick Forré,David Krueger,Joar Skalsearxiv(2024)引用 0|浏览13暂无评分AI 理解论文溯源树样例生成溯源树,研究论文发展脉络Chat Paper正在生成论文摘要