按年被提到
20192021 · 1 次2026
强化学习与智能体›博弈与游戏 AI·arXiv:1901.10995 · cs.LG

Go-Explore: a New Approach for Hard-Exploration Problems

Adrien Ecoffet, Joost Huizinga, Joel Lehman, Kenneth O. Stanley, Jeff Clune · 2019
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2021-01-05 · 最近 2026-01-01
2不同的人
2活跃月份
2讨论串
274被引(OpenAlex)
arXiv 摘要(候选层不写导读)

A grand challenge in reinforcement learning is intelligent exploration, especially when rewards are sparse or deceptive. Two Atari games serve as benchmarks for such hard-exploration domains: Montezuma's Revenge and Pitfall. On both games, current RL algorithms perform poorly, even those with intrinsic motivation, which is the dominant method to improve performance on hard-exploration domains. To address this shortfall, we introduce a new algorithm called Go-Explore. It exploits the following principles: (1) remember previously visited states, (2) first return to a promising state (without exploration), then explore from it, and (3) solve simulated environments through any available means (including by introducing determinism), then robustify via imitation learning. The combined effect of these principles is a dramatic performance improvement on hard-exploration problems. On Montezuma's Revenge, Go-Explore scores a mean of over 43k points, almost 4 times the previous state of the art. Go-Explore can also harness human-provided domain knowledge and, when augmented with it, scores a mean of over 650k points on Montezuma's Revenge. Its max performance of nearly 18 million surpasses the human world record, meeting even the strictest definition of "superhuman" performance. On Pitfall, Go-Explore with domain knowledge is the first algorithm to score above zero. Its mean score of almost 60k points exceeds expert human performance. Because Go-Explore produces high-performing demonstrations automatically and cheaply, it also outperforms imitation learning work where humans provide solution demonstrations. Go-Explore opens up many new research directions into improving it and weaving its insights into current RL algorithms. It may also enable progress on previously unsolvable hard-exploration problems in many domains, especially those that harness a simulator during training (e.g. robotics).

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论FlyingSaucer · 2021-01

“This topic is extremely interesting, and good to see that experiments support it. In AI lens: In a way, you can compare this to novelty seeking and intelligent exploration which is quite an active field in Artificial Life and game AI[1]. If you find this interesting: Jeff Clune, Kenneth Stanley and Joel Lehman conducted interesting related research. Al…”

Evidence for entropy maximisation in human free choice behaviour原评论 ↗
评论kywch · 2026-01

“I think Go-Explore (https://arxiv.org/abs/1901.10995) is promising. It'll provide automatic scaffolding and prevent catastrophic forgetting. If one can frame the problem into a competition, then self-play has been shown to work repeatedly.”

Scaffolding to Superhuman: How Curriculum Learning Solved 2048 and Tetris原评论 ↗