Yasuhiko Nakashima 研究室

主宰者：Yasuhiko Nakashima

奈良先端科学技術大学院大学

AI 要約（直近 5 年の研究成果）

本研究室は、人工知能（AI）モデルの実行を効率的にするハードウェアアクセラレータの設計と応用に関する研究を行っています。特に、大規模言語モデルや画像処理、グラフニューラルネットワークといった計算量の多い機械学習モデルを、限られた電力資源で高速に動作させることを目指しています。汎用プロセッサではなく、特定の計算に特化した再構成可能なハードウェア設計により、エネルギー効率と処理性能のバランスを追求しています。具体的には、メモリとプロセッシング要素を線形に配置した「粗粒度再構成配列」という独自のアーキテクチャを開発し、これをFPGAやASIC化することで実装しています。このハードウェア基盤の上で、大規模言語モデルの推論、医療画像処理、暗号化処理、量子回路シミュレーションなど多様なAIアプリケーションを効率的に実行する技術を開発しています。また、モデルの軽量化、メモリアクセスの最適化、計算カーネルのカスタマイズなど、ソフトウェア面での工夫も組み合わせることで、エッジデバイスやデータセンターでの実用的なAI実行環境を実現しようとしています。

※ AI（Claude）が、公開されている論文要旨から研究の問い・手法・主要な発見を事実情報として抽出・再構成して自動生成しています。誤りを含む可能性があるため、正確性は研究室公式情報でご確認ください。

外部リンク

研究成果（100 件）

[2026] A Protocol-Aware P4 Pipeline for MQTT Security and Anomaly Mitigation in Edge IoT Systems
DOI: https://doi.org/10.1109/icoin68469.2026.11480559
[2026] Spikceiver: Compact Audio-Visual Spiking Transformer with Depthwise Splitting and Bottleneck Attention
DOI: https://doi.org/10.1109/coolchips68842.2026.11556949
[2026] MergeSlide: Continual Model Merging and Task-to-Class Prompt-Aligned Inference for Lifelong Learning on Whole Slide Images
DOI: https://doi.org/10.1109/wacv61042.2026.00472
[2025] Performance Evaluation of Flan-T5 in CGLA based Accelerators
DOI: https://doi.org/10.1109/socc66126.2025.11235382
[2025] A DNN-Oriented Rapid Simulation Framework for Digital Processing-In-Memory Architectures
DOI: https://doi.org/10.1109/socc66126.2025.11235475
[2025] LLM Performance Bottlenecks on CGLA
DOI: https://doi.org/10.1109/socc66126.2025.11235453
[2025] Energy-Efficient FlashAttention Acceleration on CGLA
DOI: https://doi.org/10.1109/socc66126.2025.11235318
[2025] Efficient Hardware Implementation of Robust Binary Neural Networks Using End-to-End Unipolar Representation
DOI: https://doi.org/10.1587/transfun.2025vlp0002
[2025] An FPGA Accelerator for Vision Transformer with Quantization and LUT-Based Operations
DOI: https://doi.org/10.1587/transinf.2025pap0003
[2025] QEA: An Accelerator for Quantum Circuit Simulation with Resources Efficiency and Flexibility
DOI: https://doi.org/10.1109/icdv66179.2025.11135229

続きを表示（残り 90 件）

[2025] Correlation Power Analysis of Pipelined and Multi-Threaded Coarse-Grained Reconfigurable Cryptographic Accelerator
DOI: https://doi.org/10.1109/icdv66179.2025.11135225
[2025] A Fast and Memory-Efficient CNN Accelerator for ECG Classification in Remote Healthcare Systems
DOI: https://doi.org/10.1109/ecti-con64996.2025.11100734
[2025] UniCrypt: Universal Crypto Engine with Optimized Resource Sharing for Security Applications
DOI: https://doi.org/10.1109/coolchips65488.2025.11018585
[2025] A Lightweight Transformer Model with Dynamic Sparse Mask for Neural Machine Translation
DOI: https://doi.org/10.1109/coolchips65488.2025.11018582
[2025] MINA: A Hardware-Efficient and Flexible Mini-InceptionNet Accelerator for ECG Classification in Wearable Devices
DOI: https://doi.org/10.1109/tcsi.2025.3553837
[2025] FQsun: A Configurable Wave Function-Based Quantum Emulator for Power-Efficient Quantum Simulations
DOI: https://doi.org/10.1109/access.2025.3573313
[2025] Energy-Efficient SpMM Kernels for GATs and GCNs on a CGLA
DOI: https://doi.org/10.1109/icmlas64557.2025.10968638
[2025] Neuromorphic system using capacitor synapses
DOI: https://doi.org/10.1038/s41598-025-87924-6
[2025] Benchmarking Variants of the Adam Optimizer for Quantum Machine Learning Applications
DOI: https://doi.org/10.1109/ojcs.2025.3586953
[2025] Energy Consumption Optimization of Multi-Dimensional U-Nets on CGLA
DOI: https://doi.org/10.1109/access.2025.3539417
[2025] Optimizing Matrix-Vector Operations With CGLA for High-Performance Approximate k-NN Search
DOI: https://doi.org/10.1109/access.2025.3582825
[2025] Lifelong Whole Slide Image Analysis: Online Vision-Language Adaptation and Past-to-Present Gradient Distillation
DOI: https://doi.org/10.1109/access.2025.3580470
[2025] Efficient Kernel Mapping and Comprehensive System Evaluation of LLM Acceleration on a CGLA
DOI: https://doi.org/10.1109/access.2025.3636266
[2025] Power Efficient Attention Acceleration on CGLA
DOI: https://doi.org/10.1109/socc66126.2025.11235334
[2025] Energy-Efficient Llama3 Acceleration on a CGLA by Offloading Computational Kernels
DOI: https://doi.org/10.1109/socc66126.2025.11235407
[2024] Fusion synapse by memristor and capacitor for spiking neuromorphic systems
DOI: https://doi.org/10.1016/j.neucom.2024.127792
[2024] Power-Efficient Acceleration of GCNs on Coarse-Grained Linear Arrays
DOI: https://doi.org/10.1109/coolchips61292.2024.10531184
[2024] MRCA: Multi-grained Reconfigurable Cryptographic Accelerator for Diverse Security Requirements
DOI: https://doi.org/10.1109/coolchips61292.2024.10531185
[2024] ECG Captioning with Prior-Knowledge Transformer and Diffusion Probabilistic Model
DOI: https://doi.org/10.1007/s41666-024-00176-3
[2024] PESA: Power-Efficient SPHINCS+ Accelerator for Multi-Domain Security Applications on FPGA SoC
DOI: https://doi.org/10.1109/candar64496.2024.00037
[2024] IMAX: A Power-Efficient Multilevel Pipelined CGLA and Applications
DOI: https://doi.org/10.1109/access.2024.3524415
[2024] Efficient Random Quantum Circuit Generator: A Benchmarking Approach for Quantum Simulators
DOI: https://doi.org/10.1109/rivf64335.2024.11009047
[2024] Kyberator: A High-Efficiency FPGA-Based Multi-Mode CRYSTALS-Kyber Accelerator for Quantum-Resistant Security Applications
DOI: https://doi.org/10.1109/isocc62682.2024.10762592
[2024] High-Efficiency RISC-V-Based Cryptographic Coprocessor for Security Applications
DOI: https://doi.org/10.1109/isocc62682.2024.10762017
[2024] Theoretical Analysis of the Memory-Efficient Matrix Storage Method for Quantum Emulation Accelerators with Gate Fusion on FPGAs
DOI: https://doi.org/10.1109/mcsoc64144.2024.00067
[2024] Hator: A High-Efficiency CGRA-Based 32/64-Bit Hashing Accelerator with Real-Time Performance Analysis
DOI: https://doi.org/10.1109/mcsoc64144.2024.00033
[2024] Exploring the Limitations of Kolmogorov-Arnold Networks in Classification: Insights to Software Training and Hardware Implementation
DOI: https://doi.org/10.1109/candarw64572.2024.00026
[2024] Implementation and Evaluation of LLM on a CGLA
DOI: https://doi.org/10.1109/candar64496.2024.00040
[2024] An FPGA-Oriented Quantization Approach for Vision Transformer with LUT-Friendly Operations
DOI: https://doi.org/10.1109/candar64496.2024.00042
[2024] Benchmarking Classical and Quantum Optimizers for Quantum Simulator
DOI: https://doi.org/10.1109/candar64496.2024.00038
[2024] MRCA 2.0: An Area-Optimized Multigrained Reconfigurable Cryptographic Accelerator for Securing Blockchain-Based Internet of Things Systems
DOI: https://doi.org/10.1109/mm.2024.3501313
[2024] CTFE: A High-Efficient Heterogeneous Cryptographic CGRA for Diverse Security Applications
DOI: https://doi.org/10.1109/tcad.2024.3490176
[2024] Efficient Parameter-Shift Rule Implementation for Computing Gradient on Quantum Simulators
DOI: https://doi.org/10.1109/atc63255.2024.10908246
[2024] LI-RV: A Fast and Efficient RISC-V based Coprocessor for Lightweight Cryptography
DOI: https://doi.org/10.1109/isocc62682.2024.10762699
[2024] CGLA: Coarse-Grained Linear Array for Multi-Hash Acceleration in Blockchain Mining
DOI: https://doi.org/10.1109/isocc62682.2024.10762376
[2024] UCP: A Unified Cryptographic Processor for High Performance and Low Power Security Applications
DOI: https://doi.org/10.1109/isocc62682.2024.10762612
[2024] Quantum Battery Optimization through Quantum Machine Learning Techniques
DOI: https://doi.org/10.1109/isocc62682.2024.10762673
[2024] SuperSIM: a comprehensive benchmarking framework for neural networks using superconductor Josephson devices
DOI: https://doi.org/10.1088/1361-6668/ad6d9e
[2024] LiCryptor: High-Speed and Compact Multi-Grained Reconfigurable Accelerator for Lightweight Cryptography
DOI: https://doi.org/10.1109/tcsi.2024.3434686
[2024] HyperNTT: A Fast and Accurate NTT/INTT Accelerator with Multi-Level Pipelining and an Improved K2-RED Module
DOI: https://doi.org/10.1109/itc-cscc62988.2024.10628429
[2024] A Quantum Circuit Design for Quantum Portfolio Optimization Problem
DOI: https://doi.org/10.1109/itc-cscc62988.2024.10628188
[2024] Analog Memcapacitor by Ferroelectric Capacitor and Its Application to Spiking Neuromorphic System
DOI: https://doi.org/10.1109/ted.2024.3408783
[2024] A Fully-Parallel Reconfigurable Spiking Neural Network Accelerator with Structured Sparse Connections
DOI: https://doi.org/10.1109/iscas58744.2024.10558156
[2023] Small-Footprint Reconfigurable Heterogeneous Cryptographic Accelerator for Fog Computing
DOI: https://doi.org/10.1109/rivf60135.2023.10471770
[2023] High-efficiency Reconfigurable Crypto Accelerator Utilizing Innovative Resource Sharing and Parallel Processing
DOI: https://doi.org/10.1109/mcsoc60832.2023.00090
[2023] Universal 32/64-bit CGRA for Lightweight Cryptography in Securing IoT Data Transmission
DOI: https://doi.org/10.1109/mcsoc60832.2023.00069
[2023] A Compressed Spiking Neural Network Onto a Memcapacitive In-Memory Computing Array
DOI: https://doi.org/10.1109/mm.2023.3285529
[2023] Flexible and Scalable BLAKE/BLAKE2 Coprocessor for Blockchain-Based IoT Applications
DOI: https://doi.org/10.1109/mdat.2023.3276936
[2023] Efficient and High-Speed CGRA Accelerator for Cryptographic Applications
DOI: https://doi.org/10.1109/candar60563.2023.00033
[2023] RHCP: A Reconfigurable High-efficient Cryptographic Processor for Decentralized IoT Platforms
DOI: https://doi.org/10.1109/kse59128.2023.10299457
[2023] Versatile Resource-shared Cryptographic Accelerator for Multi-Domain Applications
DOI: https://doi.org/10.1109/icicdt59917.2023.10332303
[2023] Power-Efficient and Programmable Hashing Accelerator for Massive Message Processing
DOI: https://doi.org/10.1109/socc58585.2023.10256890
[2023] A Non-deterministic Training Approach for Memory-Efficient Stochastic Neural Networks
DOI: https://doi.org/10.1109/socc58585.2023.10256838
[2023] Time-domain Subtractive Readout Scheme for Scalable Capacitive Analog In-Memory Computing
DOI: https://doi.org/10.1109/socc58585.2023.10256918
[2023] Energy-Efficient Unified Multi-Hash Coprocessor for Securing IoT Systems Integrating Blockchain
DOI: https://doi.org/10.1109/mwscas57524.2023.10405932
[2023] Neuromorphic System Using Crosspoint-Type TaO x /Ta Memristors and Direct Device Training for Associative Memory
DOI: https://doi.org/10.1109/ted.2023.3296393
[2023] Sensitivity Analysis of Memory Bandwidth on Column-superposed Versatile Linear CGRA
DOI: https://doi.org/10.1109/newcas57931.2023.10198161
[2023] An Ultra-Compact Calculation Unit with Temporal-Spatial Re-configurability
DOI: https://doi.org/10.1109/newcas57931.2023.10198176
[2023] Phase-Change Memory using Cu2GeTe3 and Multiple Writing Technique for Neuromorphic Systems
DOI: https://doi.org/10.1109/icce56470.2023.10043395
[2022] GAND-Nets: Training Deep Spiking Neural Networks with Ternary Weights
DOI: https://doi.org/10.1109/socc56010.2022.9908132
[2022] A Coarse Grained Reconfigurable Architecture for SHA-2 Acceleration
DOI: https://doi.org/10.1109/ipdpsw55747.2022.00117
[2022] A Stochastic Coding Method of EEG Signals for Sleep Stage Classification
DOI: https://doi.org/10.1109/socc56010.2022.9908121
[2022] Application and Evaluation of Quantization for Narrow Bit-width Resampling of Sequential Monte Carlo
DOI: https://doi.org/10.1109/socc56010.2022.9908085
[2022] A Flexible and Energy-Efficient BLAKE-256/2s Co-Processor for Blockchain-based IoT Applications
DOI: https://doi.org/10.1109/sbcci55532.2022.9893257
[2022] CSIP: A Compact Scrypt IP design with single PBKDF2 core for Blockchain mining
DOI: https://doi.org/10.1109/sbcci55532.2022.9893217
[2022] Bisection Neural Network Toward Reconfigurable Hardware Implementation
DOI: https://doi.org/10.1109/tnnls.2022.3195821
[2022] A Memcapacitive Spiking Neural Network with Circuit Nonlinearity-aware Training
DOI: https://doi.org/10.1109/coolchips54332.2022.9772674
[2022] Neuromorphic chip integrated with a large-scale integration circuit and amorphous-metal-oxide semiconductor thin-film synapse devices
DOI: https://doi.org/10.1038/s41598-022-09443-y
[2022] A High-Efficiency FPGA-Based Multimode SHA-2 Accelerator
DOI: https://doi.org/10.1109/access.2022.3146148
[2022] Compact Message Permutation for a Fully Pipelined BLAKE-256/512 Accelerator
DOI: https://doi.org/10.1109/access.2022.3181410
[2022] Application of Machine Learning to Environmental DNA Metabarcoding
DOI: https://doi.org/10.1109/access.2022.3207173
[2022] Multilayer Crossbar Array of Amorphous Metal-Oxide Semiconductor Thin Films for Neuromorphic Systems
DOI: https://doi.org/10.1109/jeds.2022.3203364
[2021] XNOR-BSNN: In-Memory Computing Model for Deep Binarized Spiking Neural Network
DOI: https://doi.org/10.1109/hpbdis53214.2021.9658467
[2021] Analysis of Fully-Pipelined CNN Implementation on FPGA and HBM2
DOI: https://doi.org/10.1109/candarw53999.2021.00029
[2021] Preliminary Evaluation for Multi-domain Spike Coding on Memcapacitive Neuromorphic Circuit
DOI: https://doi.org/10.1109/candarw53999.2021.00026
[2021] Speeding Up of CGRAs by Reshaping and Stochastic FMA
DOI: https://doi.org/10.1109/candarw53999.2021.00052
[2021] Flexible Bayesian Inference by Weight Transfer for Robust Deep Neural Networks
DOI: https://doi.org/10.1587/transinf.2021edp7046
[2021] Neuromorphic System Using Memcapacitors and Autonomous Local Learning
DOI: https://doi.org/10.1109/tnnls.2021.3106566
[2021] DiaNet: An elastic neural network for effectively re-configurable implementation
DOI: https://doi.org/10.1016/j.neucom.2021.08.059
[2021] BCA: A 530-mW Multicore Blockchain Accelerator for Power-Constrained Devices in Securing Decentralized Networks
DOI: https://doi.org/10.1109/tcsi.2021.3102618
[2021] Exploring Versatility of Primary Visual Cortex Inspired Feature Extraction Hardware Model through Various Network Architectures
DOI: https://doi.org/10.1109/iccece52344.2021.9534841
[2021] An Accurate and Compact Hyperbolic Tangent and Sigmoid Computation Based Stochastic Logic
DOI: https://doi.org/10.1109/mwscas47672.2021.9531838
[2021] MuGRA: A Scalable Multi-Grained Reconfigurable Accelerator Powered by Elastic Neural Network
DOI: https://doi.org/10.1109/tcsi.2021.3099034
[2021] Development of Spiking Neural Network Using Mem Capacitor
[2021] Neuromorphic System with Thin-Film Memdevices
[2021] Amorphous-Metal-Oxide-Semiconductor Thin-Film Planar-Type Spike-Timing- Dependent-Plasticity Synapse Device
DOI: https://doi.org/10.1109/led.2021.3082083
[2021] Training Low-Latency Spiking Neural Network through Knowledge Distillation
DOI: https://doi.org/10.1109/coolchips52128.2021.9410323
[2021] High Performance Multicore SHA-256 Accelerator using Fully Parallel Computation and Local Memory
DOI: https://doi.org/10.1109/coolchips52128.2021.9410349
[2021] Neuromorphic System with Thin-Film 3D-Integrated Devices and Local Automatic Learning
[2021] Amorphous metal oxide semiconductor thin film, analog memristor, and autonomous local learning for neuromorphic systems
DOI: https://doi.org/10.1038/s41598-020-79806-w

科研費（0 件）

まだデータがありません（KAKEN 取り込み後に表示）。

所属学会・役職（0 件）

まだデータがありません（学会データ連携後に表示）。

AI 要約（直近 5 年の研究成果）

外部リンク

関連研究室(8 件)

研究成果（100 件）

科研費（0 件）

所属学会・役職（0 件）