AIMR Math-AI seminar #S2


Date : Feb. 6th (Wed.) 14:00~16:00

Place:3C, AIMR Main Bldg., Katahira campus, Tohoku University

Speaker:Dr. Yoshiki Sugitani

Title:Training of neural networks with biased datasets and its application to anemia control system


Training in machine learning is mathematically nothing but solving the minimization problem of a nonlinear loss function.
However, in practice you need some ingenuity in order to obtain high generalization performance. Although a larger amount of training data makes more success of machine learning, it is often difficult to prepare enough datasets in real world problems.
In this talk, some techniques to handle the classification problems with biased datasets are introduced.And as an application, we report the result of an anemia control system developed for dialysis treatment.


Speaker:Dr. Edgar Avalos

Title:Classification of x-ray images of epoxy resins


Epoxy resins used in the aerospace industry contain a considerable variety of inhomogeneities. Correctly identifying the role of such inhomogeneities on the material properties is an extraordinary challenge. We propose a methodology to categorize samples of different types of epoxy resins. Basic statistical quantities suggest that the first two principal components are related to the mean and variance of detrended data. Some methods of classification based on machine learning are presented.




【Latest News】

Math discussion for ATP

Date: April 19th (Fri) 2019
Speaker: Dr. Hirofumi Oka
Speaker: Prof. Tomoteru Fukumura

【More Info】


AIMR Math Group Seminar

Date: April 4th (Thu) 2019 
Speaker: Dr. Koji Sato

【More Info】






















 Advanced Institute for Materials Research Tohoku University Mathematical Group
 2-1-1 Katahira, Aoba-ku, Sendai, 980-8577 Japan 【Access/Contact