Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations

TL;DR

This paper introduces a novel task inversion learning framework that enables robots to generalize manipulation skills to new conditions using inverse task learning from forward demonstrations, improving zero-shot transfer accuracy.

Contribution

The authors propose a joint learning approach that constructs a shared representation for forward and inverse tasks, enabling effective knowledge transfer without direct supervision.

Findings

Outperforms diffusion-based methods in complex manipulation tasks

Demonstrates successful zero-shot generalization in real-world experiments

Shows robustness across diverse objects and tools

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.