# Compressing the Backward Pass of Large-Scale Neural Architectures by Structured Activation Pruning

Source: [Compressing the Backward Pass of Large-Scale Neural Architectures by Structured Activation Pruning](<https://www.sota2.com/research/paper/68fa715e1e209131cea50e91>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:04:39.575Z

Authors: Daniel Barley, Holger Fr\\"oning

Published: 2023-11-28T15:31:31.000Z

## Abstract

The rise of Deep Neural Networks (DNNs) has led to an increase in model size and complexity, straining the memory capacity of GPUs. Sparsity in DNNs, characterized as structural or ephemeral, has gained attention as a solution. This work focuses on ephemeral sparsity, aiming to reduce memory consumption during training. It emphasizes the significance of activations, an often overlooked component, and their role in memory usage. This work employs structured pruning in Block Sparse Compressed Row (BSR) format in combination with a magnitude-based criterion to efficiently prune activations. We furthermore introduce efficient block-sparse operators for GPUs and showcase their effectiveness, as well as the superior compression offered by block sparsity. We report the effectiveness of activation pruning by evaluating training speed, accuracy, and memory usage of large-scale neural architectures on the example of ResMLP on image classification tasks. As a result, we observe a memory reduction of up to 32% while maintaining accuracy. Ultimately, our approach aims to democratize large-scale model training, reduce GPU requirements, and address ecological concerns.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2311.16883>)

[PDF](<https://arxiv.org/pdf/2311.16883.pdf>)
