LAB LOGA page on LabLog, a review site (not the lab's official page)
Katagiri Lab (High-Performance Computing and Auto-Tuning)
Machine-translated · View original
A lab that connects generative AI with high-performance computing.
AboutAbout the lab
Toward high-performance computing written by humans and AI together.
The Katagiri-Hoshino Lab studies generative AI, numerical computation, and software auto-tuning, with high-performance computing (HPC) at its core.
Its research addresses the shift from "HPC where humans write everything" to "HPC where AI and humans collaboratively design, generate, verify, and optimize."
Using supercomputers, the lab works on massively parallel computing, large-scale machine learning, GPU computing, and quantum-computer-related technologies.
Undergraduate students in the School of Informatics, Department of Computer Science can list this lab as a preference when they are assigned to labs in their fourth year.
Key terms
- High-Performance Computing (HPC)
- A technical field concerned with performing large-scale, time-consuming computations quickly using supercomputers and similar systems.
- AI agent
- AI that advances toward a goal on its own, for example by writing and running programs. In this lab, multiple agents are assigned different roles.
- Auto-tuning
- Technology that automatically selects algorithms, code, and parameters according to the computer and input, so as to maintain performance.
- Performance portability
- The ability to maintain high execution performance even when the computing environment changes.
NewsLab news
Established the Agentic Computing Research Unit (ACRU)
Presented research on online reinforcement learning for HPC code generation at LLM4HPC 2026
Presented a paper on VibeCodeHPC at iWAPT 2026
Received the Best Paper Award at LLM4HPCAsia 2026
Released VibeCodeHPC, which coordinates multiple AI coding CLIs
ResearchResearch themes
Centered on high-performance computing, the lab conducts research across numerical algorithms, software, AI, and computer architecture. It is extending conventional auto-tuning technology to generative AI and AI agents, expanding into research in which AI supports the whole process from understanding the requirements of HPC code through generation, verification, and performance evaluation.
*Illustrative imageGenerative AI, AI Agents, and HPC
Research on making autonomous the LLM-based generation of HPC code in C, Fortran, CUDA, and other languages, as well as performance optimization, evaluation on real machines, debugging, and verification. It also covers role division among multiple agents, local LLMs, and knowledge accumulation.
*Illustrative imageSoftware Auto-Tuning
Research on automatically selecting algorithms, code, and parameters according to the architecture and input in order to achieve performance portability. It covers auto-tuning languages, performance models, search, mixed precision, and runtime optimization.
*Illustrative imageFast Implementation on GPUs and Massively Parallel Systems
Research on fast implementations on GPUs, multi-core and many-core processors, and supercomputers that take into account the characteristics of computation, memory, and communication. Targets include CUDA, OpenMP, OpenACC, and MPI.
*Illustrative imageNumerical Linear Algebra and High-Precision Computing
Covers eigenvalue and singular value problems, LU/QR decomposition, Krylov subspace methods, GMRES, BiCGStab, preconditioning, mixed precision, and verified numerical computation, along with improving their performance.
MembersFaculty
- ProfessorTakahiro Katagiri
- Associate ProfessorTetsuya Hoshino
- Assistant ProfessorDaichi Mukunoki
VoicesStudent voices
Ratings and reviews reflect the experiences of individual current or former students, not the lab's official views. Whether a lab has reviews, or how many, says nothing about the lab itself.
Students who take part in an interview may receive a thank-you gift worth ¥1,000, regardless of what they say.
Student ratings (0 students)
--
--
--
--
--
--
Student voices answer questions like these
- What's great about this lab?
- What did you learn after joining?
- What's the vibe, and what is the professor like?
- Who is this lab for?
- Where do graduates go?
Your experience can help younger students choose their lab. Your name won't be shown.
Write a lab reviewAnonymous · Your draft is saved if you stopRelatedLabs similar to Katagiri Lab (High-Performance Computing and Auto-Tuning)
Shimada Lab (Information Network Systems)School of Informatics, Department of Computer ScienceEducation and research on the construction of information ne…
Murase LabSchool of Informatics, Department of Computer ScienceEducation and research on the construction of information ne…
Kaji LabSchool of Informatics, Department of Computer ScienceEducation and research, from both theoretical and practical …
Embedded Real-Time Systems Lab (ERTL)School of Informatics, Department of Computer ScienceEducation and research on platforms realizing highly reliabl…The information on this page is compiled by the LabLog team from the lab's public information and interviews. For corrections or removal requests, please contact LabLog, not the lab.
Report an errorAbout the images
The top and research theme images are illustrative and were not taken at the lab. They are cropped from Wikimedia Commons images. Cropped CC BY-SA images are provided under the same CC BY-SA license as the original (version shown on each line).
- Top image: “Nanoscience High-Performance Computing Facility” Matt Howard / CC BY-SA 2.0 (cropped)
- Generative AI, AI Agents, and HPC: “Pleiades supercomputer racks 2” Oleg Alexandrov / CC BY-SA 4.0 (cropped)
- Software Auto-Tuning: “Programming code” Martin Vorel / CC BY-SA 4.0 (cropped)
- Fast Implementation on GPUs and Massively Parallel Systems: “Oak Ridge - Kraken (Cray XT5)” Daderot / Public domain (cropped)
- Numerical Linear Algebra and High-Precision Computing: “DEC PDP-10 (DECsystem 10) mainframe computer system at Living Computer Museum (strongly edited image)” Joe Mabel / CC BY-SA 3.0 (cropped)
- Related lab (Shimada Lab (Information Network Systems)): “DOME-microDataCenterFull” RonaldGlider / CC BY-SA 4.0 (cropped)
- Related lab (Murase Lab): “networks, Tomaszów Mazowiecki, Poland” WrS.tm.pl / CC0 (cropped)
- Related lab (Kaji Lab): “Centaur server room ( )” VIA Gallery from Hsintien, Taiwan / CC BY 2.0 (cropped)
- Related lab (Embedded Real-Time Systems Lab (ERTL)): “Electronic Component” Electronicspices / CC BY-SA 4.0 (cropped)
About LabLog
LabLog is a lab review app for students only. Student members at each university interview students in labs and publish what they say. Student voices can only be read by students who sign up with a university email address. Operated by bestiee Inc.
Learn more about LabLog ↗Search engines and AI may reference the basic lab info and rating summaries on this page and the description of LabLog. Student voices are for student members only. Automated collection by programs (scraping, or input to or training of AI) is prohibited under Article 8 of the Terms of Service. If you need data for research or analysis, please contact us.