Wavelet Policy: Lifting Scheme for Policy Learning in Long-Horizon Tasks

TL;DR

This paper introduces a wavelet-based policy learning framework that employs multi-scale wavelet analysis to improve decision-making in long-horizon, complex tasks for embodied AI systems.

Contribution

It presents a novel wavelet policy method using learnable multi-scale decomposition and lifting schemes for enhanced policy learning in extended sequences.

Findings

Improved policy precision in robotic manipulation tasks.

Enhanced robustness in self-driving scenarios.

Effective multi-robot collaboration demonstrated.

Abstract

Policy learning focuses on devising strategies for agents in embodied artificial intelligence systems to perform optimal actions based on their perceived states. One of the key challenges in policy learning involves handling complex, long-horizon tasks that require managing extensive sequences of actions and observations with multiple modes. Wavelet analysis offers significant advantages in signal processing, notably in decomposing signals at multiple scales to capture both global trends and fine-grained details. In this work, we introduce a novel wavelet policy learning framework that utilizes wavelet transformations to enhance policy learning. Our approach leverages learnable multi-scale wavelet decomposition to facilitate detailed observation analysis and robust action planning over extended sequences. We detail the design and implementation of our wavelet policy, which incorporates lifting schemes for effective multi-resolution analysis and action generation. This framework is evaluated across multiple complex scenarios, including robotic manipulation, self-driving, and multi-robot collaboration, demonstrating the effectiveness of our method in improving the precision and reliability of the learned policy.