拡散ポリシーの各構成要素を分解して検証する
Unpacking the Individual Components of Diffusion Policy
模倣学習手法Diffusion Policyの5つの主要構成要素(観測系列入力、行動系列実行、後退ホライズン、U-Net/Transformer、FiLM条件付け)をManiSkillとAdroitで実験し、各要素の寄与を明らかにした。
著者: Xiu Yuan
分類: cs.LG, cs.RO
原文アブストラクト
Imitation Learning presents a promising approach for learning generalizable and complex robotic skills. The recently proposed Diffusion Policy generates robot action sequences through a conditional denoising diffusion process, achieving state-of-the-art performance compared to other imitation learning methods. This paper summarizes five key components of Diffusion Policy: 1) observation sequence input; 2) action sequence execution; 3) receding horizon; 4) U-Net or Transformer network architecture; and 5) FiLM conditioning. By conducting experiments across ManiSkill and Adroit benchmarks, this study aims to elucidate the contribution of each component to the success of Diffusion Policy in various scenarios. We hope our findings will provide valuable insights for the application of Diffusion Policy in future research and industry.
