AutoSculpt: A Pattern-based Model Auto-pruning Framework Using Reinforcement Learning and Graph Learning

TL;DR

AutoSculpt introduces a pattern-based auto-pruning framework for DNNs that leverages graph learning and reinforcement learning to optimize model compression and runtime efficiency.

Contribution

It presents a novel pattern-based pruning method that automatically identifies and prunes regular structures in DNNs using graph encoding and DRL, improving efficiency and accuracy.

Findings

Achieves up to 90% pruning rates across various architectures.

Nearly 18% FLOPs reduction compared to baseline methods.

Outperforms existing auto-pruning approaches in accuracy and efficiency.

Abstract

As deep neural networks (DNNs) are increasingly deployed on edge devices, optimizing models for constrained computational resources is critical. Existing auto-pruning methods face challenges due to the diversity of DNN models, various operators (e.g., filters), and the difficulty in balancing pruning granularity with model accuracy. To address these limitations, we introduce AutoSculpt, a pattern-based automated pruning framework designed to enhance efficiency and accuracy by leveraging graph learning and deep reinforcement learning (DRL). AutoSculpt automatically identifies and prunes regular patterns within DNN architectures that can be recognized by existing inference engines, enabling runtime acceleration. Three key steps in AutoSculpt include: (1) Constructing DNNs as graphs to encode their topology and parameter dependencies, (2) embedding computationally efficient pruning patterns, and (3) utilizing DRL to iteratively refine auto-pruning strategies until the optimal balance between compression and accuracy is achieved. Experimental results demonstrate the effectiveness of AutoSculpt across various architectures, including ResNet, MobileNet, VGG, and Vision Transformer, achieving pruning rates of up to 90% and nearly 18% improvement in FLOPs reduction, outperforming all baselines. The codes can be available at https://github.com/jlx15588/AutoSculpt