Computer Architecture Research Group

Our laboratory pioneers the future of integrated circuits that underpin intelligent systems across modern society. As AI technology evolves at an explosive pace, conventional computing systems are reaching a major turning point. We approach this from two directions — hardware for AI (chips that run AI fast and efficiently) and AI for hardware (using AI to automate complex chip design with EDA tools) — and aim to vertically integrate these into real robotic systems. From theoretical algorithm research to FPGA and ASIC implementation and real-robot control, we foster world-class computer engineers through end-to-end research that bridges software and hardware.

Architecture

Beyond mere computer organization, this encompasses system design that includes the very environment in which robots operate. We view LSIs (integrated circuits) not as plain computing platforms but as the core of embodied intelligence that interacts with the physical world.

Robotics

This is the process of embodying the intelligence embedded in circuits into a physical entity. We implement state-of-the-art AI techniques such as imitation learning on our own ultra-low-power on-chip learning platform.

Computing

We pursue the question of how to efficiently design and build the very foundation of computation — the integrated circuit. Currently, our focus is on next-generation EDA environments driven by LLM-agent-based design automation.

News & Topics

More