Rewarded meta-pruning: Meta Learning with Rewards for Channel Pruning

TL;DR

This paper introduces Rewarded meta-pruning, a novel meta-learning approach with reward functions to optimize channel pruning in CNNs, balancing accuracy and efficiency for deployment on resource-constrained devices.

Contribution

It proposes a reward-based meta-learning algorithm for channel pruning that effectively controls the trade-off between accuracy and computational efficiency.

Findings

Outperforms state-of-the-art pruning methods on ResNet-50, MobileNetV1, and MobileNetV2.

Effectively balances accuracy and efficiency through accuracy and efficiency coefficients.

Demonstrates significant reduction in parameters and FLOPs with maintained accuracy.

Abstract

Convolutional Neural Networks (CNNs) have a large number of parameters and take significantly large hardware resources to compute, so edge devices struggle to run high-level networks. This paper proposes a novel method to reduce the parameters and FLOPs for computational efficiency in deep learning models. We introduce accuracy and efficiency coefficients to control the trade-off between the accuracy of the network and its computing efficiency. The proposed Rewarded meta-pruning algorithm trains a network to generate weights for a pruned model chosen based on the approximate parameters of the final model by controlling the interactions using a reward function. The reward function allows more control over the metrics of the final pruned model. Extensive experiments demonstrate superior performances of the proposed method over the state-of-the-art methods in pruning ResNet-50, MobileNetV1, and MobileNetV2 networks.