この検索内の頻出タグ: dnn^:6 accelerators^:3 energy^:3 inference^:3 8-bit^:2 fault^:2 injections^:2 input^:2 model^:2 multiplier^:2 parameters^:2 segmentation^:2 squeezing^:2 vulnerability^:2 adaptive^:1 all^:1 bit-width^:1 energy-efficient^:1 estimating^:1 estimation^:1 few^:1 minimizing^:1 multipliers^:1 number^:1 s3m^:1 scaling^:1 semi-segmented^:1 small^:1 static^:1 training^:1

6 件の該当がありました． : このページのURL : HTML

論文誌

[1] Y. Zhang, H. Itsuji, T. Uezono, T. Toba, and M. Hashimoto, "Vulnerability Estimation of DNN Model Parameters with Few Fault Injections," IEICE Trans. on Fundamentals of Electronics, Communications and Computer Sciences, volume E106-A, number 3, pages 523-531, March 2023. [pdf]

国際会議

[1] M. Zhang and M. Hashimoto, "Squeezing 8-Bit Multiplier Energy with Input Segmentation in DNN Inference Accelerators," International collaboration Symposium on Information, Production and Systems (ISIPS), November 2024.

[2] M. Zhang, Q. Cheng, H. Awano, L. Lin, and M. Hashimoto, "S3M: Static Semi-Segmented Multipliers for Energy-Efficient DNN Inference Accelerators," Proceedings of IEEE International Conference on Computer Design (ICCD), pages 16-23, October 2024. [pdf]

[3] Y. Zhang, H. Itsuji, T. Uezono, T. Toba, and M. Hashimoto, "Estimating Vulnerability of All Model Parameters in DNN with a Small Number of Fault Injections," Proceedings of Design, Automation and Test in Europe Conference (DATE), pages 60-63, March 2022. [pdf]

[4] T. Cheng and M. Hashimoto, "Minimizing Energy of DNN Training with Adaptive Bit-Width and Voltage Scaling," Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS), May 2021. [pdf]

国内会議(査読付き)

[1] Zhang Mingtao, Cheng Quan, 粟野 皓光, Lin Longyang, 橋本 昌宜, "Squeezing 8-Bit Multiplier Energy with Input Segmentation in DNN Inference Accelerators," 情報処理学会DAシンポジウム, 2024年.