ピクセル入力連続制御における視覚的汎化の測定
Measuring Visual Generalization in Continuous Control from Pixels
既存の連続制御タスクに多様な視覚変化を加えたベンチマークを提案し、現在の強化学習手法の視覚的汎化性能を評価した。データ拡張が自己教師あり学習より優れ、強い画像変換が汎化に有効であることを示した。
著者: Jake Grigsby, Yanjun Qi
分類: cs.LG, cs.AI, cs.CV, cs.RO
原文アブストラクト
Self-supervised learning and data augmentation have significantly reduced the performance gap between state and image-based reinforcement learning agents in continuous control tasks. However, it is still unclear whether current techniques can face a variety of visual conditions required by real-world environments. We propose a challenging benchmark that tests agents' visual generalization by adding graphical variety to existing continuous control domains. Our empirical analysis shows that current methods struggle to generalize across a diverse set of visual changes, and we examine the specific factors of variation that make these tasks difficult. We find that data augmentation techniques outperform self-supervised learning approaches and that more significant image transformations provide better visual generalization \footnote{The benchmark and our augmented actor-critic implementation are open-sourced @ https://github.com/QData/dmc_remastered)
