用对称结构约束潜在空间，让模型更好理解环境变化规律。
Learning Abstract World Models with a Group-Structured Latent Space
- 用群结构设计潜在空间，融入旋转平移等对称性先验
- 在3D环境中预测更准，下游强化学习性能提升明显
- 生成更简洁、解耦的抽象表示，适合复杂动态系统建模
学习马尔可夫决策过程（MDP）的有意义抽象模型对于从有限数据中实现良好泛化至关重要。本文提出通过几何先验约束学习到的转移模型低维表示流形。通过合理选择潜在空间及其关联的群作用，引入已知对称结构，编码环境中不变性的先验知识。同时，框架允许将额外非结构化信息与这些对称性一同嵌入。实验表明，相比完全无结构的方法，该方法在具有旋转和平移特征的环境中（包括3D环境的第一人称视角）能获得更优的潜在转移模型预测，且在下游强化学习任务中表现更佳。此外，该方法生成的表示更简单、更解耦。完整代码已开源以确保可复现性。
原文摘要 · Abstract (English)
Learning meaningful abstract models of Markov Decision Processes (MDPs) is crucial for improving generalization from limited data. In this work, we show how geometric priors can be imposed on the low-dimensional representation manifold of a learned transition model. We incorporate known symmetric structures via appropriate choices of the latent space and the associated group actions, which encode prior knowledge about invariances in the environment. In addition, our framework allows the embedding of additional unstructured information alongside these symmetries. We show experimentally that this leads to better predictions of the latent transition model than fully unstructured approaches, as well as better learning on downstream RL tasks, in environments with rotational and translational features, including in first-person views of 3D environments. Additionally, our experiments show that this leads to simpler and more disentangled representations. The full code is available on GitHub to ensure reproducibility.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。