量子化不要の自己回帰アクショントランスフォーマ
Quantization-Free Autoregressive Action Transformer
行動空間を離散化せず、連続的な自己回帰トランスフォーマで模倣学習を行う手法を提案し、模擬ロボットタスクで最高性能を達成した。
著者: Ziyad Sheebaelhamd, Michael Tschannen, Michael Muehlebach, Claire Vernade
分類: cs.LG, cs.RO
原文アブストラクト
Current transformer-based imitation learning approaches introduce discrete action representations and train an autoregressive transformer decoder on the resulting latent code. However, the initial quantization breaks the continuous structure of the action space thereby limiting the capabilities of the generative model. We propose a quantization-free method instead that leverages Generative Infinite-Vocabulary Transformers (GIVT) as a direct, continuous policy parametrization for autoregressive transformers. This simplifies the imitation learning pipeline while achieving state-of-the-art performance on a variety of popular simulated robotics tasks. We enhance our policy roll-outs by carefully studying sampling algorithms, further improving the results.
