End-to-End自動運転のメカニスティック解釈可能性
Mechanistic Interpretability for End-to-End Self-Driving
単眼カメラのEnd-to-End経路計画モデルの内部活性にSparse Autoencoderを適用し、経路意図や速度・ブレーキ挙動に対応する解釈可能な特徴を抽出。特徴操作で計画出力が予測的に変化することを示し、SAE特徴を用いたアダプタで軌道選択誤差を3.44%削減した。
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: Manav Gagvani, Benjamin Namikas, Sivamurugan Velmurugan, Nysa Kumar, Shrey Sharma, Thomas Peterson, Pratyush Mathur, Hyunseo Chang, Rohan Paleja
分類: cs.RO, cs.CV, cs.LG
原文アブストラクト
End-to-end (E2E) autonomous driving models achieve strong planning performance but remain largely opaque, limiting observability and trust. We apply mechanistic interpretability to E2E driving by training Sparse Autoencoders (SAEs) on the internal activations of a simple, single-camera end-to-end path planning model. Unsupervised clustering of SAE latents reveals interpretable structure aligned with route intent, vehicle speed, and braking behavior, recovering route intent with 98.3% accuracy without supervision. Through paired counterfactual image edits, we further show that specific SAE features are visually grounded and systematically respond to scene changes such as traffic light state and stop sign presence. Steering individual SAE latents produces predictable, monotonic changes in planner outputs including lateral displacement, curvature, and braking. Finally, we demonstrate a practical application of these representations: we build a lightweight adapter trained on SAE latents that reduces trajectory selection error by 3.44%, outperforming an equivalent adapter that omits SAE latents, confirming that the sparse representation captures useful information for interpretability. Code and visualizations are available at https://mgagvani.github.io/mechanistic-interpretability-for-e2e-driving/