LAB LOGA page on LabLog, a review site (not the lab's official page)
Kanamori Lab
Machine-translated · View original
A lab that rethinks machine learning and statistics from a mathematical perspective.
AboutAbout the lab
Exploring the essence of information through machine learning and statistics.
The Kanamori Lab studies machine learning and mathematical statistics.
Machine learning and statistical techniques that dig out and make use of the treasure hidden in data are having a major impact on society.
The lab aims to understand information quantitatively from a mathematical perspective and to apply the results across data science in general.
It also works on information geometry and on optimization methods for learning algorithms.
Key terms
- Machine learning
- A technology that finds patterns in large amounts of data and uses them to make predictions and classifications for new data.
- Mathematical statistics
- A field of mathematics that reads out the properties of data containing variability, based on probability theory.
- Information geometry
- An approach that views a collection of probability distributions as a geometric shape and studies the mechanisms of statistics and learning in the language of geometry.
ResearchResearch themes
The lab focuses on machine learning and mathematical statistics, aiming to understand information quantitatively from a mathematical perspective. It also studies information geometry and optimization methods for learning algorithms.
*Illustrative imageLearning even with few correct labels
In machine learning, collecting data with correct labels takes a lot of effort. Based on statistical theory, the lab studies methods that can learn correctly even from data in which only some of the data have correct labels, or in which only the mixing proportions are known.
*Illustrative imageStatistical estimation robust to outliers
When outliers or noise get mixed into data, learning results can go badly wrong. By devising ways to measure the distance between distributions, the lab develops estimation methods that are robust to outliers, and also applies them to training generative models.
*Illustrative imageKnowing how reliable AI predictions are
To judge whether an AI's prediction can be trusted as it is, it is important to know how reliable that prediction is. The lab is developing methods that estimate uncertainty by slightly perturbing a trained model, as well as techniques for investigating the phenomenon in which understanding suddenly advances partway through training.
MembersFaculty
- ProfessorTakafumi KanamoriPublications ↗
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 Kanamori Lab
Takabe LabSchool of Computing, Department of Mathematical and Computing ScienceStatistical physics; information processing by machine learn…
Nanashima LabSchool of Computing, Department of Mathematical and Computing ScienceComputational complexity theory, learning, and cryptography …
Sakamoto LabSchool of Computing, Department of Mathematical and Computing ScienceHigh-performance, low-power computers
Takahashi LabSchool of Computing, Department of Mathematical and Computing ScienceSingularities of solutions to nonlinear partial differential…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: “T-SNE Embedding of MNIST” Kyle McDonald / CC BY 2.0 (cropped)
- Learning even with few correct labels: “A trainer instructing a data annotator on how to label images by Nacho Kamenov & Humans in the Loop” Nacho Kamenov and Humans in the Loop / CC BY 4.0 (cropped)
- Statistical estimation robust to outliers: “Scatter plot showing positive correlation with outliers” Monika Wahi / CC BY-SA 4.0 (cropped)
- Knowing how reliable AI predictions are: “Gaussianprocess gapUncertainty” Physikinger / CC0 (cropped)
- Related lab (高邉賢史研究室): “Ising-tartan” Linas / CC BY-SA 3.0 (cropped)
- Related lab (七島幹人研究室): “Turing Machine Model Davey 2012” Rocky Acosta / CC BY 3.0 (cropped)
- Related lab (坂本龍一研究室): “Racks Amravati Data Center” PiDatacenters / CC BY-SA 4.0 (cropped)
- Related lab (高橋仁研究室): “Heat” Nicoguaro. Based on File:Heat eqn.gif by en:User:Oleg Alexandrov / CC BY 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.