LAB LOGA page on LabLog, a review site (not the lab's official page)
Statistical Intelligence Lab
Machine-translated · View original
A lab covering statistics and machine learning, from theory to applications.
AboutAbout the lab
Searching for the laws behind data through mathematics
The Statistical Intelligence Lab (Shimodaira-Honda Lab) studies the methodology of statistics and machine learning, from theory to applications.
Statistics plays an important role as the theoretical foundation supporting data science and artificial intelligence.
In research on theory and methods, the lab emphasizes mathematics and programming, and in applications it works with natural language processing and biological data.
Key terms
- Language model
- A model that uses probability to express which words are likely to come next in a text. It is at the core of generative AI
- KL divergence
- A measure of how far apart two probability distributions are
- Bandit problem
- A problem of continuing to choose among options whose probabilities of winning are unknown, trying them out so as to maximize the total reward
NewsLab news
View More ↗ResearchResearch themes
The lab studies statistics and machine learning from theory to applications. Topics include language models and embeddings, bandit problems, Bayesian optimization, graph embedding, and the bootstrap and selective inference.
*Illustrative imageA map of generative AI models
Mathematically, a generative AI model is a probability distribution. Using the ideas of information geometry, the lab computed coordinates that approximate the deviation between models (KL divergence), and proposed a method that arranges more than 1,000 language models like a map.
*Illustrative imageBandit problems: choosing while trying
The lab studies the problem of choosing the best action while collecting data, without sufficient data in advance. For bandit problems, which are used in recommender systems and elsewhere, it theoretically clarifies the limits of what can be achieved and builds practical algorithms.
*Illustrative imageWord embeddings and statistical inference
The lab studies the shape common to language and image embeddings, the strength of meaning represented by embedding length, and analogies such as drug and gene in word embeddings. It also covers inference on complex networks, as well as the bootstrap and selective inference.
MembersFaculty
The faculty of the Statistical Intelligence Lab are Professor Hidetoshi Shimodaira and Associate Professor Junya Honda.
- ProfessorHidetoshi Shimodaira
- Associate ProfessorJunya Honda
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 Statistical Intelligence Lab
Computer Software LabFaculty of Engineering, School of Informatics and Mathematical ScienceResearch on theory and applications for building highly effi…
Collective Intelligence Systems LabFaculty of Engineering, School of Informatics and Mathematical ScienceCollective intelligence systems: advancing fundamental techn…
Mathematical Information Systems LabFaculty of Engineering, School of Informatics and Mathematical ScienceEducation and research on the mathematical problems needed t…
Agreement Informatics LabFaculty of Engineering, School of Informatics and Mathematical ScienceAI-powered collective intelligence in which people, computer…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: “Normal distribution function” Shishirdasika / CC BY-SA 3.0 (cropped)
- A map of generative AI models: “4 continuous probability density functions” Ldecola / CC0 (cropped)
- Bandit problems: choosing while trying: “Slot machines at McCarran International Airport” Grendelkhan / CC BY-SA 4.0 (cropped)
- Word embeddings and statistical inference: “CBOW eta Skipgram” Aelu013 / CC BY-SA 4.0 (cropped)
- Related lab (Computer Software Lab): “Source code in Javascript” Romainhk / CC BY-SA 3.0 (cropped)
- Related lab (Collective Intelligence Systems Lab): “DataAppliedWeb” Marksenizer / CC BY-SA 3.0 (cropped)
- Related lab (Mathematical Information Systems Lab): “Cosine Series Plus Noise” Fourier1789 / CC BY-SA 4.0 (cropped)
- Related lab (Agreement Informatics Lab): “Discussion Between National Consensus Commission And Nationalist Likeminded Alliance (PID-0002082)” Press Information Department / Public domain (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.