Pruning-based Data Selection and Network Fusion for Efficient Deep Learning

TL;DR

PruneFuse is a novel method that combines pruning and network fusion to select informative data samples efficiently and accelerate deep neural network training, reducing computational costs and improving performance.

Contribution

The paper introduces PruneFuse, a new approach that integrates pruning and network fusion for efficient data selection and faster training in deep learning.

Findings

Reduces computational costs for data selection

Achieves better performance than baseline methods

Speeds up overall network training process

Abstract

Efficient data selection is essential for improving the training efficiency of deep neural networks and reducing the associated annotation costs. However, traditional methods tend to be computationally expensive, limiting their scalability and real-world applicability. We introduce PruneFuse, a novel method that combines pruning and network fusion to enhance data selection and accelerate network training. In PruneFuse, the original dense network is pruned to generate a smaller surrogate model that efficiently selects the most informative samples from the dataset. Once this iterative data selection selects sufficient samples, the insights learned from the pruned model are seamlessly integrated with the dense model through network fusion, providing an optimized initialization that accelerates training. Extensive experimentation on various datasets demonstrates that PruneFuse significantly reduces computational costs for data selection, achieves better performance than baselines, and accelerates the overall training process.