Check the preview of 2nd version of this platform
being developed by the open MLCommons taskforce on automation and reproducibility
as a free, open-source and technology-agnostic on-prem platform.
GPU Memory Management for Deep Neural Networks Using Deep Q-Network
lib:da111b4f3db30bea (v1.0.0)
Authors: Anonymous
Where published: ICLR 2020 1
Document: PDF DOI
Abstract URL: https://openreview.net/forum?id=BJxg7eHYvB
Deep neural networks use deeper and broader structures to achieve better performance and consequently, use increasingly more GPU memory as well. However, limited GPU memory restricts many potential designs of neural networks. In this paper, we propose a reinforcement learning based variable swapping and recomputation algorithm to reduce the memory cost, without sacrificing the accuracy of models. Variable swapping can transfer variables between CPU and GPU memory to reduce variables stored in GPU memory. Recomputation can trade time for space by removing some feature maps during forward propagation. Forward functions are executed once again to get the feature maps before reuse. However, how to automatically decide which variables to be swapped or recomputed remains a challenging problem. To address this issue, we propose to use a deep Q-network(DQN) to make plans. By combining variable swapping and recomputation, our results outperform several well-known benchmarks.
Where published: ICLR 2020 1
Document: PDF DOI
Abstract URL: https://openreview.net/forum?id=BJxg7eHYvB
Deep neural networks use deeper and broader structures to achieve better performance and consequently, use increasingly more GPU memory as well. However, limited GPU memory restricts many potential designs of neural networks. In this paper, we propose a reinforcement learning based variable swapping and recomputation algorithm to reduce the memory cost, without sacrificing the accuracy of models. Variable swapping can transfer variables between CPU and GPU memory to reduce variables stored in GPU memory. Recomputation can trade time for space by removing some feature maps during forward propagation. Forward functions are executed once again to get the feature maps before reuse. However, how to automatically decide which variables to be swapped or recomputed remains a challenging problem. To address this issue, we propose to use a deep Q-network(DQN) to make plans. By combining variable swapping and recomputation, our results outperform several well-known benchmarks.
Relevant initiatives
Related knowledge about this paper
Reproduced results (crowd-benchmarking and competitions)
Artifact and reproducibility checklists
- ACM/cTuning artifact appendix and reproducibility checklist
- NeurIPS reproducibility checklist
- SIGPLAN's checklist for empirical evaluation
- Collective Knowledge (organizing research projects based on FAIR principles)