身体化AIのための加法的合成潜在行動の学習
Learning Additively Compositional Latent Actions for Embodied AI
視覚遷移から潜在行動を学習する際に、物理運動の加法的・合成的構造を強制するモデルAC-LAMを提案し、下流のポリシー学習の性能を向上させた。
詳しい要約
1. どんなもの？
2. 先行研究と比べてどこがすごい？
3. 技術・手法の肝は？
4. どうやって有効だと検証した？
5. 議論はある？
6. 次に読むべき論文は？
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Hangxing Wei, Xiaoyu Chen, Chuheng Zhang, Tim Pearce, Jianyu Chen, Alex Lamb, Li Zhao, Jiang Bian
分類: cs.CV, cs.AI
原文アブストラクト
Latent action learning infers pseudo-action labels from visual transitions, providing an approach to leverage internet-scale video for embodied AI. However, most methods learn latent actions without structural priors that encode the additive, compositional structure of physical motion. As a result, latents often entangle irrelevant scene details or information about future observations with true state changes and miscalibrate motion magnitude. We introduce Additively Compositional Latent Action Model (AC-LAM), which enforces scene-wise additive composition structure over short horizons on the latent action space. These AC constraints encourage simple algebraic structure in the latent action space~(identity, inverse, cycle consistency) and suppress information that does not compose additively. Empirically, AC-LAM learns more structured, motion-specific, and displacement-calibrated latent actions and provides stronger supervision for downstream policy learning, outperforming state-of-the-art LAMs across simulated and real-world tabletop tasks.
