Research map / Reinforcement learning / Imitation learning

RAIL: Representation-Aligned Imitation Learning for Student-Compatible Teacher Policies

NeurIPS 2026 · Reinforcement learning › Imitation learning · approach: Imitation combined with RL

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes RAIL, a representation-aligned imitation learning method that makes privileged teacher policies compatible with raw-sensory student policies for distillation.

Problem

RL from raw sensory inputs (images, onboard sensors) is hard due to high-dimensional observations and sparse rewards; distilling a privileged-state teacher into a raw-input student fails when the teacher relies on information the student cannot access, making exact imitation problematic.

Approach

Aligns the representations of the student with the teacher so the teacher becomes student-compatible, rather than forcing exact action imitation; details not available beyond the abstract.

Closest papers by meaning

• Student-Informed Teacher Training (ICLR 2025 · similarity 0.92)
• Rethinking Inverse Reinforcement Learning: from Data Alignment to Task Alignment (NeurIPS 2024 · similarity 0.87)
• Multi-Teacher Knowledge Distillation with Reinforcement Learning for Visual Recognition (AAAI 2025 · similarity 0.86)
• Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model (ICLR 2025 · similarity 0.85)
• Off-Dynamics Reinforcement Learning via Domain Adaptation and Reward Augmented Imitation (NeurIPS 2024 · similarity 0.85)
• Mimicking the Maestro: Exploring the Efficacy of a Virtual AI Teacher in Fine Motor Skill Acquisition (AAAI 2024 · similarity 0.85)