Keyframe-Focused Visual Imitation Learning

Understand

Imitation learning trains control policies by mimicking pre-recorded expert demonstrations.

• In partially observable settings, imitation policies must rely on observation histories, but many seemingly paradoxical results show better performance for policies that only access the most recent observation.
• Recent solutions ranging from causal graph learning to deep information bottlenecks have shown promising results, but failed to scale to realistic settings such as visual imitation.
• We propose a solution that outperforms these prior approaches by upweighting demonstration keyframes corresponding to expert action changepoints.