敵対的スキルネットワーク：動画からの教師なしロボットスキル学習
Adversarial Skill Networks: Unsupervised Robot Skill Learning from Video
ラベルなしの多視点動画から、タスクに依存しないスキル埋め込み空間を敵対的損失とメトリック学習で学習し、新規タスクへのスキル転用・組み合わせを可能にする手法を提案した論文。
著者: Oier Mees, Markus Merklinger, Gabriel Kalweit, Wolfram Burgard
分類: cs.CV, cs.LG, cs.RO
原文アブストラクト
Key challenges for the deployment of reinforcement learning (RL) agents in the real world are the discovery, representation and reuse of skills in the absence of a reward function. To this end, we propose a novel approach to learn a task-agnostic skill embedding space from unlabeled multi-view videos. Our method learns a general skill embedding independently from the task context by using an adversarial loss. We combine a metric learning loss, which utilizes temporal video coherence to learn a state representation, with an entropy regularized adversarial skill-transfer loss. The metric learning loss learns a disentangled representation by attracting simultaneous viewpoints of the same observations and repelling visually similar frames from temporal neighbors. The adversarial skill-transfer loss enhances re-usability of learned skill embeddings over multiple task domains. We show that the learned embedding enables training of continuous control policies to solve novel tasks that require the interpolation of previously seen skills. Our extensive evaluation with both simulation and real world data demonstrates the effectiveness of our method in learning transferable skills from unlabeled interaction videos and composing them for new tasks. Code, pretrained models and dataset are available at http://robotskills.cs.uni-freiburg.de
