フィッシャー保存ガイダンス:安全な拡散制御のための学習不要の多様体制約
Fisher-Preserving Guidance: Training-Free Manifold Constraints for Safe Diffusion Control
拡散モデルによる軌道予測で、学習データの多様体から外れる更新を防ぐ学習不要の推論手法を提案し、ナビゲーションや操作タスクで性能を向上させた。
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: Hao Ren, Zetong Bi, Yiming Zeng, Le Zheng, Zhi Li, Zhaoliang Wan, Lu Qi, Hui Cheng
分類: cs.RO, cs.LG
原文アブストラクト
Diffusion models are effective for waypoint prediction in visual navigation, but standard sampling and test time guidance can produce unreliable or inefficient trajectories when updates drift off the training manifold. We propose Fisher Preserving Guidance with Outer Product Span Projection, a training-free inference method that avoids large Fisher drift associated with off-distribution actions while optimizing a task objective. Our method computes the Fisher-preserving update via a low-rank Jacobian factorization, requiring only a single backward pass per step and enabling real-time use. We further introduce Truncated Fisher Denoising Sensitivity as an uncertainty signal and use it for robust multi-sample action blending. Experiments on toy and realistic navigation benchmarks, including Maze2D with TSDF-based guidance, PushT with official Diffusion Policy weights, and visual navigation in simulation and on real robots, demonstrate consistent improvements in performance over strong diffusion-policy baselines without additional training.
