BriefGPT.xyz
Jun, 2024
TEDi策略：时域纠缠扩散用于机器人控制
TEDi Policy: Temporally Entangled Diffusion for Robotic Control
HTML
PDF
Sigmund H. Høeg, Lars Tingelstad
TL;DR
使用Temporally Entangled Diffusion (TEDi)框架，改进基于扩散的策略提取和训练，在保持性能的同时大幅提升采样速度。
Abstract
diffusion models
have been shown to excel in
robotic imitation learning
by mastering the challenge of modeling complex distributions. However,
sa
→