Energy-efficient XNOR-free In-Memory BNN Accelerator with Input Distribution Regularization
- Title
- Energy-efficient XNOR-free In-Memory BNN Accelerator with Input Distribution Regularization
- Authors
- HYUNGJUN, KIM; OH, HYUNMYUNG; KIM, JAE JOON
- Date Issued
- 2020-11-03
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Abstract
- SRAM-based in-memory Binary Neural Network (BNN) accelerators are garnering interests as a platform for energy-efficient edge neural network computing thanks to their compactness in terms of hardware and neural network parameter size. However, previous works had to modify SRAM cells to support XNOR operations on memory array resulting in limited area and energy efficiencies. In this work, we present a conversion method which replaces the signed inputs (+1/-1) of BNN with the unsigned inputs (1/0) without computation error, and vice versa. The method enables BNN computing on conventional 6T SRAM arrays and improves area and energy efficiencies. We also demonstrate that further energy saving is possible by skewing the distribution of binary input data based on regularization during network training. Evaluation results show that the proposed techniques improve the inference energy efficiency by up to 9.4x for various benchmarks over previous works.
- URI
- https://oasis.postech.ac.kr/handle/2014.oak/105720
- Article Type
- Conference
- Citation
- 39th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2020, 2020-11-03
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.