Resarch
Research Area
Computer Architecture
Processor Architecture and Memory System
Hardware Security and Secure Processor
Hardware Design Technology and High-Level Synthesis Compiler
Algorithm/Hardware Co-design
Machine Learning System, Distributed Machine Learning, and Federated Learning
FPGA System
We are pursuing various research on highly efficient next-generation computers, such as custom computing employing FPGA and domain-specific hardware, algorithm/hardware co-design for machine learning, and high-level synthesis compilers for a productive hardware design environment. We are also pursuing software research on compilers, programming models, and frameworks for user/programmer-friendly computers.
Recent Research Projects
Federated Learning for Dynamic and Diverse Environments
Communication-Efficient Federated Learning
Security of Federated Learning (Backdoor Attack and Protection)
Personalized Federated Learning
Large-Scale Federated Learning Simulator and Platform
Decision Tree Ensemble based Federated Learning
Online Learning Algorithm/Accelerator
Domain-Specific Architecture and Machine Learning Accelerator
DNN Training/Inference Accelerator (AI Chip)
Hardware-Oriented DNN Algorithm (Quantized Neural Network, Binary Neural Network)
Computation-Efficient Bayes Deep Learning Algorithm and Hardware
Model Compression Algorithm for Hard-wired DNN Accelerators
Hardware/Algorithm Co-design for Decision Tree Ensemble
Accelerator for Multimodal AI and Robotics
Computer Architecture
Secure Processor (Trusted Execution Environment; TEE)
Memory Encryption and Integrity Verification
Processing in Memory Circuit, Processor, and Compiler
Energy-Efficient GPU Architecture
Approximate Computing Architecture based on Stochastic Thermodynamics
Hardware Design Technology and Compiler
Hardware Design Language and High-Level Synthesis Compiler
Python-based Hardware Design Framework Veriloggen
DNN Accelerator Synthesis Compiler NNgenÂ
Hardware Description Obfuscation for IP Protection
Covert Channel in Hardware Systems