LAB LOGA page on LabLog, a review site (not the lab's official page)
Takeuchi Lab
Machine-translated · View original
A lab that builds computers in which data processing and memory are fused, as in the brain.
AboutAbout the lab
Toward computers that fuse data processing and memory
The Takeuchi Lab studies data-centric computing, in which data processing and memory are fused as in the brain.
In image and speech recognition, the lab also researches approaches that tolerate a certain amount of error in order to achieve fast processing and a substantial reduction in power consumption.
The goal is to realize computers that can process and store, in real time, huge volumes of data that change moment by moment.
The lab trains people who can "Co-design" across fields, from LSI hardware to control software, machine learning, and applications.
Key terms
- CiM
- Short for Computation in Memory. A technology that performs computation inside memory, reducing data movement and power consumption
- ReRAM
- Non-volatile memory that stores information through changes in electrical resistance. Data is retained even when the power is turned off
- Adversarial attack
- An attack that adds small alterations, imperceptible to humans, to data in order to deliberately make an AI reach a wrong decision
NewsLab news
Won the Outstanding Presentation Award at the IPSJ SLDM Special Interest Group meeting
A paper was published in IEICE Transactions on Information and Systems
Won the Outstanding Poster Award at DesignGaia 2025
Won the Best Student Award at AEC/APC Symposium Asia 2025
Succeeded in achieving both large capacity through multi-level storage and 10-year data retention in ReRAM CiM (announced together with Nuvoton)
ResearchResearch themes
The lab studies data-centric computing in which data processing and memory are fused as in the brain. By co-designing hardware and software, it aims to build intelligent computers that see, hear, and think like humans.
*Illustrative imageLong-lasting multi-level ReRAM CiM
This research uses multi-level ReRAM, which stores multiple values in a single device, for CiM, which performs computation inside memory. The lab is developing a mechanism that compensates for the errors caused by memory degradation over time, keeping the inference accuracy of AI intact even after 10 years or more.
*Illustrative imagePhysical AI
AI that enables robots and autonomous vehicles to perceive the real world and act on their own as conditions change. The lab combines Liquid Neural Networks, inspired by neural circuits, with low-power hardware such as CiM using ReRAM.
*Illustrative imageDiffusion language models
Language models that apply diffusion models, used in image generation, to text generation, producing multiple words at the same time. The lab is developing technology to reduce memory and power use by efficiently placing the large KV cache required during generation in memory such as ReRAM.
*Illustrative imageAI robust against adversarial attacks
This research aims to increase resistance to adversarial attacks, which deliberately cause AI to malfunction. By using the computation errors caused by variation in CiM memory during training, the lab makes AI robust against attacks without additional hardware.
MembersFaculty
The members of the Takeuchi Lab consist of faculty, secretaries, academic specialists, and doctoral, master's, and undergraduate students.As of 2026
- ProfessorKen TakeuchiPublications ↗
- Project Associate ProfessorChihiro MatsuiPublications ↗
VoicesStudent voices
Compiled from answers by students who were in the lab. Ratings and reviews reflect individual students' experiences, not the lab's official views.
Students who take part in an interview may receive a thank-you gift worth ¥1,000, regardless of what they say.
Who is this lab for?
What's great about this lab?
What did you learn after joining?
What's the vibe, and what is the professor like?
Where do graduates go?
RelatedLabs similar to Takeuchi Lab
Iizuka LabFaculty of Engineering, Department of Information and Communication EngineeringDigital–analog mixed-signal circuits and systems in ultra-sc…
Irie-Kadomoto LabFaculty of Engineering, Department of Information and Communication EngineeringResearch centered on processor architecture, encompassing la…
Kosuge Lab (Klab)Faculty of Engineering, Department of Information and Communication EngineeringLow-power, high-performance semiconductor integrated circuit…
Kawahara LabFaculty of Engineering, Department of Information and Communication EngineeringAI-enabled digital manufacturing technology, wireless power …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: “Datacenter Server Racks ( )” Carl Lender / CC BY 2.0 (cropped)
- Long-lasting multi-level ReRAM CiM: “Spiky Nickel Nanowire networks” Saharalialy / CC BY-SA 4.0 (cropped)
- Physical AI: “Ouster OS1-64 lidar point cloud of intersection of Folsom and Dore St, San Francisco” Daniel L. Lu / CC BY 4.0 (cropped)
- Diffusion language models: “Diffusion with ink” Zvonimir Lončarić / CC BY-SA 4.0 (cropped)
- AI robust against adversarial attacks: “Static viewed on a cathode ray tube television from close up” Boreal11 / CC0 (cropped)
- Related lab (Iizuka Lab): “07R01” Mister rf / CC BY-SA 4.0 (cropped)
- Related lab (Irie-Kadomoto Lab): “12-inch silicon wafer” Peellden / CC BY-SA 3.0 (cropped)
- Related lab (Kosuge Lab (Klab)): “Intel@14nm@Skylake@Skylake-SP(XCC)@Xeon(ES)@QJW5 DSCx8 poly@20x - Flickr - FritzchensFritz” Fritzchens Fritz from Berlin / CC0 (cropped)
- Related lab (Kawahara Lab): “Printed flexible electronics ( )” Oak Ridge National Laboratory / CC BY 2.0 (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.