Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations

2026-03-05

AI Summary

This work addresses the limited out-of-distribution generalization of robotic policies, the fragility of imitation learning, and the heavy reliance of transfer learning on large amounts of annotated data. To overcome these challenges, the authors propose a joint learning framework that requires no inverse-task supervision signals. By constructing a shared representation between forward and inverse tasks and leveraging unsupervised forward demonstrations from novel configurations to guide inverse-task learning, the method achieves, for the first time, efficient and accurate zero-shot inference of inverse policies using only forward demonstrations. Evaluated on simulated and real-world multi-object, multi-tool manipulation tasks, the approach significantly outperforms diffusion-model-based baselines, demonstrating strong cross-domain extrapolation capabilities.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment or tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges by framing the problem in the context of task inversion learning and proposing a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We show the extrapolation capabilities of our framework via ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based alternatives.

Problem

• task generalization
• zero-shot extrapolation
• robot learning
• imitation learning
• transfer learning

Innovation

• task inversion learning
• zero-shot generalization
• forward-inverse task representation
• imitation learning
• policy extrapolation