Session Details

[24p-31A-1~14]FS.1 Focused Session "AI Electronics"

Sun. Mar 24, 2024 1:00 PM - 5:00 PM JST
Sun. Mar 24, 2024 4:00 AM - 8:00 AM UTC
31A (Building No. 3)
Takao Marukame(Toshiba), Takahide Oya(Yokohama Natl. Univ.)

[24p-31A-1][INVITED] Advanced Flash Memory Technologies and its Application to Neuromorphic Computing

〇Yuichiro Mitani1 (1.Tokyo City Univ.)

[24p-31A-2]Modeling of Inherent Noise of Floating Gate (FG) Technology-based Stochastic Neurons and Effects of Noise on Signal Detection Accuracy in Stochastic Resonance

〇Akira Goda1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-3]Effects of Threshold Voltage Variation and Electron Injection Noise of FG Technology based Stochastic Neurons on Stochastic Resonance Characteristics

〇Akira Goda1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-4]Performance Improvement of Deep Reinforcement Learning in Computation-in-Memory by Quantization-Aware & Write Variation-Aware Training

〇(B)Ryugo Sato1, Kenshin Yamauchi1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-5]Co-design of Highly Error Tolerant FeFET-based CiM and Strong Lottery Ticket Hypothesis

〇(M1)Kenshin Yamauchi1, Ayumu Yamada1, Naoko Misawa1, Seong-Kun Cho1, Kasidit Toprasertpong1, Shinichi Takagi1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-6]Classification Method of ReRAM Current Fluctuation by CNN and Physical Model of Fluctuation

〇Ayumu Yamada1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-7]Circuit Design of FeFET-based Voltage-sensing Computation-in-Memory

〇Chihiro Matsui1, Kasidit Toprasertpong1, Shinichi Takagi1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-8]Towards Small-Area, Low-Energy and Highly Accurate CiM I: I/O Range Training Method

〇Ayumu Yamada1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-9]Towards Small-Area, Low-Energy and Highly Accurate CiM II: Error Compensation

〇Ayumu Yamada1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-10]Quantization Method and Quantization Aware Training with Computation-in-Memory for Making Vision Transformer Compact

〇Naoko Misawa1, Ryuhei Yamaguchi1, Ayumu Yamada1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-11]Computation-in-Memory Integrating of Generating Random Weights & MAC Operation for Neuromorphic Computing

〇Naoko Misawa1, Shunsuke Koshino1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-12]Application of GNN and CNN to CiM-based Accelerators

〇Hanxi Xue1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)

[24p-31A-13]Analysis of Low-Bit Precision ReRAM CiM-based Convolutional Neural Networks during Training and Inference

〇(D)Adil Padiyal1, Ayumu Yamada1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.The Univ. Of Tokyo)

[24p-31A-14]Analysis of Read Current Fluctuation in Low Resistance State ReRAM by using Fluctuation Pattern Classifier

〇(M2)Zhiyuan Huang1, Ayumu Yamada1, Naoko Misawa1, Chihiro Matsui1, Ken Takeuchi1 (1.Univ. Tokyo)