paper-with-me

홈 › Papers

Exploiting Sparsity in Pruned Neural Networks to Optimize Large Model Training

2023-02-10 · Siddharth Singh, Abhinav Bhatele

Parallel training of neural networks at scale is challenging due to significant overheads arising from communication. Recently, deep learning researchers have developed a variety of pruning algorithms that are capable of pruning (i.e. setting to zero) 80-90% of the parameters in a neural network to yield sparse subnetworks that equal the accuracy of the unpruned parent network. In this work, we propose a novel approach that exploits these sparse subnetworks to optimize the memory utilization and communication in two popular algorithms for parallel deep learning namely -- data and inter-layer parallelism. We integrate our approach into AxoNN, a highly scalable framework for parallel deep learning that relies on data and inter-layer parallelism, and demonstrate the reduction in communication time and memory utilization. On 512 NVIDIA V100 GPUs, our optimizations reduce the memory consumption of a 2.7 billion parameter model by 74%, and the total communication time by 40%, thus providing an overall speedup of 34% over AxoNN, 32% over DeepSpeed-3D and 46% over Sputnik, a sparse matrix computation baseline.

📄 PDF Abstract BibTeX arXiv:2302.05045

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Learning

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

Omni-sparsity DNN: Fast Sparsity Optimization for On-Device Streaming E2E ASR via Supernet

2021-10-15 · Haichuan Yang, Yuan Shangguan, Dilin Wang, Meng Li 외

From wearables to powerful smart devices, modern automatic speech recognition (ASR) models run on a variety of edge devices with different computational budgets. To navigate the Pareto front of model accuracy vs model si…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)GPUNavigate+2

N2NSkip: Learning Highly Sparse Networks using Neuron-to-Neuron Skip Connections

2022-08-07 · Arvind Subramaniam, Avinash Sharma

The over-parametrized nature of Deep Neural Networks leads to considerable hindrances during deployment on low-end devices with time and space constraints. Network pruning strategies that sparsify DNNs using iterative pr…

Network Pruning

MACKO: Sparse Matrix-Vector Multiplication for Low Sparsity

2025-11-17 · Vladimír Macko, Vladimír Boža arxiv

Sparse Matrix-Vector Multiplication (SpMV) is a fundamental operation in the inference of sparse Large Language Models (LLMs). Because existing SpMV methods perform poorly under the low and unstructured sparsity (30-90%)…

Spartus: A 9.4 TOp/s FPGA-based LSTM Accelerator Exploiting Spatio-Temporal Sparsity

2021-08-04 · Chang Gao, Tobi Delbruck, Shih-Chii Liu

Long Short-Term Memory (LSTM) recurrent networks are frequently used for tasks involving time-sequential data such as speech recognition. Unlike previous LSTM accelerators that either exploit spatial weight sparsity or t…

speech-recognitionSpeech Recognition

The Unreasonable Effectiveness of Random Pruning: Return of the Most Naive Baseline for Sparse Training

2022-02-05 · ICLR 2022 4 · Shiwei Liu, Tianlong Chen, Xiaohan Chen, Li Shen 외

Random pruning is arguably the most naive way to attain sparsity in neural networks, but has been deemed uncompetitive by either post-training pruning or sparse training. In this paper, we focus on sparse training and hi…

Adversarial RobustnessOut-of-Distribution Detection