paper-with-me

홈 › Papers

Sparse Weight Activation Training

2020-01-07 · NeurIPS 2020 12 · Md Aamir Raihan, Tor M. Aamodt

Neural network training is computationally and memory intensive. Sparse training can reduce the burden on emerging hardware platforms designed to accelerate sparse computations, but it can affect network convergence. In this work, we propose a novel CNN training algorithm Sparse Weight Activation Training (SWAT). SWAT is more computation and memory-efficient than conventional training. SWAT modifies back-propagation based on the empirical insight that convergence during training tends to be robust to the elimination of (i) small magnitude weights during the forward pass and (ii) both small magnitude weights and activations during the backward pass. We evaluate SWAT on recent CNN architectures such as ResNet, VGG, DenseNet and WideResNet using CIFAR-10, CIFAR-100 and ImageNet datasets. For ResNet-50 on ImageNet SWAT reduces total floating-point operations (FLOPS) during training by 80% resulting in a 3.3$\times$ training speedup when run on a simulated sparse learning accelerator representative of emerging platforms while incurring only 1.63% reduction in validation accuracy. Moreover, SWAT reduces memory footprint during the backward pass by 23% to 50% for activations and 50% to 90% for weights.

📄 PDF Abstract BibTeX arXiv:2001.01969

Code (1)

AamirRaihan/SWAT 공식 구현 pytorch

Tasks

Image ClassificationNetwork PruningSparse Learning

Methods 이 논문이 사용한 방법론

Concatenated Skip Connection A Concatenated Skip Connection is a type of skip connection that seeks to reuse features by concatenating them to new layers, allowing more information to be retained from…
Batch Normalization 설명 없음
Dense Block A Dense Block is a module used in convolutional neural networks that connects *all layers* (with matching feature-map sizes) directly with each other. It was originally…
Average Pooling 설명 없음
ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 설명 없음
Max Pooling Max Pooling is a pooling operation that calculates the maximum value for patches of a feature map, and uses it to create a downsampled (pooled) feature map. It is usually…

Similar Papers 제목 키워드 기반

WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation Sparsity

2026-02-16 · Lei Chen, Yuan Meng, Xiaoyu Zhan, Zhi Wang 외 arxiv

Large Language Models (LLMs) offer strong capabilities but incur high inference costs due to dense computation and memory access. Training-free activation sparsity is a promising approach for efficient LLM inference, yet…

WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference

2025-05-26 · Sihan Chen, Dan Zhao, Jongwoo Ko, Colby Banbury 외

The growing computational demands of large language models (LLMs) make efficient inference and activation strategies increasingly critical. While recent approaches, such as Mixture-of-Experts (MoE), leverage selective ac…

Language ModelingLanguage ModellingLarge Language ModelMixture-of-Experts

Activation function impact on Sparse Neural Networks

2020-10-12 · Adam Dubowski

While the concept of a Sparse Neural Network has been researched for some time, researchers have only recently made notable progress in the matter. Techniques like Sparse Evolutionary Training allow for significantly low…

Learning Neural Networks with Sparse Activations

2024-06-26 · Pranjal Awasthi, Nishanth Dikkala, Pritish Kamath, Raghu Meka

A core component present in many successful neural network architectures, is an MLP block of two fully connected layers with a non-linear activation in between. An intriguing phenomenon observed empirically, including in…

QuEST: Stable Training of LLMs with 1-Bit Weights and Activations

2025-02-07 · Andrei Panferov, Jiale Chen, Soroush Tabesh, Roberto L. Castro 외

One approach to reducing the massive costs of large language models (LLMs) is the use of quantized or sparse representations for training or deployment. While post-training compression methods are very popular, the quest…

GPUQuantization