论文标题
云：无监督动态的对比度学习
CLOUD: Contrastive Learning of Unsupervised Dynamics
论文作者
论文摘要
可以从高维观察（例如像素）执行复杂控制任务的开发代理，这是由于有效学习动态的困难而具有挑战性。在这项工作中，我们建议通过对比度估计以完全无监督的方式学习前进和反向动态。具体来说，我们在状态的特征空间中训练一个前向动力学模型和一个逆动力学模型，并使用随机探索收集的数据进行训练。与大多数现有的确定性模型不同，我们的基于能量的模型考虑了代理环境相互作用的随机性质。我们证明了方法在各种任务中的功效，包括目标指导的计划和观察的模仿。项目视频和代码在https://jianrenw.github.io/cloud/。
Developing agents that can perform complex control tasks from high dimensional observations such as pixels is challenging due to difficulties in learning dynamics efficiently. In this work, we propose to learn forward and inverse dynamics in a fully unsupervised manner via contrastive estimation. Specifically, we train a forward dynamics model and an inverse dynamics model in the feature space of states and actions with data collected from random exploration. Unlike most existing deterministic models, our energy-based model takes into account the stochastic nature of agent-environment interactions. We demonstrate the efficacy of our approach across a variety of tasks including goal-directed planning and imitation from observations. Project videos and code are at https://jianrenw.github.io/cloud/.
