LAB LOGA page on LabLog, a review site (not the lab's official page)
Kohjima Lab
Machine-translated · View original
The lab does machine learning research on learning from imperfect real-world data.
AboutAbout the lab
Toward machine learning that can learn from imperfect data.
The Kohjima Lab studies machine learning that can learn from real-world data that is insufficient in both quantity and quality.
The lab aims to expand the areas where machine learning can be effective and to improve the performance of AI systems.
This is an NTT collaborative lab. The faculty member belongs to Nippon Telegraph and Telephone Corporation and conducts research in cooperation with Professor Minami.
For students who wish to take part, the lab would like to create opportunities to experience corporate research activities, such as internships at NTT laboratories.
Key terms
- Machine learning
- Technology that enables computers to learn from large amounts of data and automatically discover the rules and patterns hidden in it
- Transition probability
- The probability of moving from one location to another. In this lab, it is estimated from imperfect mobility data
- Regression analysis
- Analysis that captures the quantitative relationship between input variables and output variables. For example, it examines the relationship between study time and exam scores
ResearchResearch themes
This lab studies machine learning that can learn from data that is insufficient in both quantity and quality, or that comes in an imperfect form different from what is normally expected. By representing the process that generates the data with a mathematical model and deriving an algorithm for parameter estimation, the lab makes imperfect data analyzable.
*Illustrative imageLearning from imperfect mobility data
Location data can become mobility data with missing information on intermediate waypoints, because of thinning for privacy protection or areas where positioning is not possible. The lab proposed a method to estimate transition probabilities between arbitrary locations from such data, making it possible to understand human mobility trends based on the spatial structure of a city.
*Illustrative imageLearning from imperfect input-output data
When input and output information are collected by different departments or at the group level, the result is disjointed data in which the correspondence between the two is lost. The lab proposed a method to learn a regression function from such data with a deep neural network, enabling regression analysis without restrictions on the form of the true regression function.
MembersFaculty
- Affiliated Associate ProfessorMasahiro Kohjima
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 Kohjima Lab
Ueno LabSchool of Informatics and Engineering, Class I (Informatics)Artificial intelligence / Machine learning / Data science
Orihara LabSchool of Informatics and Engineering, Class I (Informatics)Machine learning, data mining, text mining
Hara LabSchool of Informatics and Engineering, Class I (Informatics)Machine learning / data science / trustworthy AI
Harada LabSchool of Informatics and Engineering, Class I (Informatics)Machine learning / data science / data mining / artificial i…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: “Machine Learning cookies” Syced / CC0 (cropped)
- Learning from imperfect mobility data: “Applications of Elk Location Data ( )” USFWS Mountain Prairie / Public domain (cropped)
- Learning from imperfect input-output data: “Feature Learning Diagram” Fgpacini / CC BY-SA 4.0 (cropped)
- Related lab (植野 真臣 研究室): “Bayesian network” Sachs K, Gifford D, Jaakkola T, Sorger P, Lauffenburger DA. / CC BY-SA 4.0 (cropped)
- Related lab (折原 良平 研究室): “Young people in Hong Kong using smartphones whilst walking” Ccmsharma2 / CC BY-SA 4.0 (cropped)
- Related lab (原 聡 研究室): “Client Project Hashcat Testing With an AMD Epyc Supermicro Nvidia GPU Server Using Proxmox (Lawrence Systems) 07” Lawrence Systems / CC BY 3.0 (cropped)
- Related lab (原田 慧 研究室): “Real-time bus tracking control room in Lebanon” UrusHyby / CC0 (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.