---
title: "Inferring Transferable Rewards via Active Inverse Reinforcement Learning"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.03296v1"
permalink: "https://feed.astralor.com/items/itm_d6fc595fd05c728a6461618e"
published: "2026-11-02T15:32:37.000Z"
score: 55
channel: "ai"
tags: ["Inferring Transferable Rewards","Active Inverse Reinforcement","Learning Inverse Reinforcement","Learning","IRL","Prior","Inverse Reinforcement Learning"]
---

# Inferring Transferable Rewards via Active Inverse Reinforcement Learning

## At a glance

Inverse Reinforcement Learning (IRL) aims to recover expert rewards that transfer across environments with differing dynamics. Prior work has shown that such transfer is possible when gathering expert demonstrations acr…

## Key points

- Inverse Reinforcement Learning (IRL) aims to recover expert rewards that transfer across environments with differing dynamics.
- Prior work has shown that such transfer is possible when gathering expert demonstrations across sufficiently different dynamics.
- In practice, however, expert demonstrations are costly, making it critical to select the right dynamics to reduce the quantity of expert data required for learning transferable rewards.

## Full text

Inverse Reinforcement Learning (IRL) aims to recover expert rewards that transfer across environments with differing dynamics. Prior work has shown that such transfer is possible when gathering expert demonstrations across sufficiently different dynamics. In practice, however, expert demonstrations are costly, making it critical to select the right dynamics to reduce the quantity of expert data required for learning transferable rewards. We propose a two-phase approach: we first cycle through the available dynamics to obtain a reliable reward estimate, then actively select dynamics to reduce our uncertainty about the expert reward. Crucially, this yields novel sample complexity guarantees on transfer that hold for any dynamics, with a rate independent of the number of available dynamics. Empirically, our method significantly reduces the number of expert demonstrations needed to learn a transferable reward.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_d6fc595fd05c728a6461618e