Implicit Curriculum in Procgen Made Explicit
Zhenxiong Tan, Kaixin Wang, Xinchao Wang
摘要
Procedurally generated environments such as Procgen Benchmark provide a testbed for evaluating the agent’s ability to robustly learn a relevant skill, by situating the agent in ever-changing levels. The diverse levels associated with varying contexts are naturally connected to curriculum learning. Existing works mainly focus on arranging the levels to explicitly form a curriculum. In this work, we take a close look at the learning process itself under the multi-level training in Procgen. Interestingly, the learning process exhibits a gradual shift from easy contexts to hard contexts, suggesting an implicit curriculum in multi-level training. Our analysis is made possible through C-Procgen , a benchmark we build upon Procgen that enables explicit control of the contexts. We believe our findings will foster a deeper understanding of learning in diverse contexts, and our benchmark will benefit future research in curriculum reinforcement learning.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了最后一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
它引用的顶会 Paper9
- Leveraging Procedural Generation to Benchmark Reinforcement LearningKarl Cobbe, Christopher Hesse, Jacob Hilton, John SchulmanICML 2020 · 被引用 685 次
- Emergent Complexity and Zero-shot Transfer via Unsupervised Environment DesignMichael Dennis, Natasha Jaques, Eugene Vinitsky, Alexandre M. Bayen 等NeurIPS 2020 · 被引用 362 次
- Prioritized Level ReplayMinqi Jiang, Edward Grefenstette, Tim RocktäschelICML 2021 · 被引用 211 次
- Evolving Curricula with Regret-Based Environment DesignJack Parker-Holder, Minqi Jiang, Michael Dennis, Mikayel Samvelyan 等ICML 2022 · 被引用 175 次
- Enhanced POET: Open-ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their SolutionsRui Wang, Joel Lehman, Aditya Rawal, Jiale Zhi 等ICML 2020 · 被引用 148 次
相关 Paper
- Adaptive Procedural Task Generation for Hard-Exploration ProblemsKuan Fang, Yuke Zhu, Silvio Savarese, Li Fei-FeiICLR 2021 · 被引用 36 次
- TeachMyAgent: a Benchmark for Automatic Curriculum Learning in Deep RLClément Romac, Rémy Portelas, Katja Hofmann, Pierre-Yves OudeyerICML 2021 · 被引用 28 次
- Curriculum Reinforcement Learning via Constrained Optimal TransportPascal Klink, Haoyi Yang, Carlo D'Eramo, Jan Peters 等ICML 2022 · 被引用 44 次
- CurBench: Curriculum Learning BenchmarkYuwei Zhou, Zirui Pan, Xin Wang, Hong Chen 等ICML 2024 · 被引用 11 次
- Diffusion-based Curriculum Reinforcement LearningErdi Sayar, Giovanni Iacca, Ozgur S. Oguz, Alois KnollNeurIPS 2024 · 被引用 12 次
