Exploiting sparsity in pruned neural networks to optimize large model training
Published in 2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2023
Recommended citation: @INPROCEEDINGS{10177389, author={Singh, Siddharth and Bhatele, Abhinav}, booktitle={2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)}, title={Exploiting Sparsity in Pruned Neural Networks to Optimize Large Model Training}, year={2023}, volume={}, number={}, pages={245-255}, keywords={Deep learning;Training;Artificial satellites;Computational modeling;Neural networks;Memory management;Parallel processing;lottery ticket hypothesis;sparse computations;GPUs;parallel deep learning;memory optimizations}, doi={10.1109/IPDPS54959.2023.00033}}
Download Paper