スコア・分布マッチング方策：マッチング蒸留による高速視覚運動方策
Score and Distribution Matching Policy: Advanced Accelerated Visuomotor Policies via Matched Distillation
拡散方策をスコアマッチングと分布マッチングの二段階最適化で1ステップ生成器に変換し、6倍の推論高速化と高品質な行動生成を両立させた。
著者: Bofang Jia, Pengxiang Ding, Can Cui, Mingyang Sun, Pengfang Qian, Siteng Huang, Zhaoxin Fan, Donglin Wang
分類: cs.RO, cs.LG, stat.ML
原文アブストラクト
Visual-motor policy learning has advanced with architectures like diffusion-based policies, known for modeling complex robotic trajectories. However, their prolonged inference times hinder high-frequency control tasks requiring real-time feedback. While consistency distillation (CD) accelerates inference, it introduces errors that compromise action quality. To address these limitations, we propose the Score and Distribution Matching Policy (SDM Policy), which transforms diffusion-based policies into single-step generators through a two-stage optimization process: score matching ensures alignment with true action distributions, and distribution matching minimizes KL divergence for consistency. A dual-teacher mechanism integrates a frozen teacher for stability and an unfrozen teacher for adversarial training, enhancing robustness and alignment with target distributions. Evaluated on a 57-task simulation benchmark, SDM Policy achieves a 6x inference speedup while having state-of-the-art action quality, providing an efficient and reliable framework for high-frequency robotic tasks.
