Futoshi Futami - Looking for students

Recruiting Students in Machine Learning

I am a member of Sugimoto Laboratory, Division of Mathematical Science, Department of Systems Science, Graduate School of Engineering Science, the University of Osaka.

Students who wish to join our laboratory are required to pass the entrance examination for either the Master’s or Doctoral Program of the Graduate School of Engineering Science. Please refer to the official admission guidelines for details.

Alternatively, undergraduate students who enroll in the Mathematical Science Course, Department of Information Science may choose Sugimoto Laboratory after admission.

If you are interested in pursuing graduate studies or conducting research with us, please feel free to contact me.

Research Topics

Our goal is to make machine learning more reliable and trustworthy for real-world applications. Our current research topics include:

Uncertainty Quantification of machine learning algorithms (e.g., calibration and decision theory) and their theoretical analysis
Statistical learning theory and algorithm design for latent variable models using deep learning methods such as Variational Autoencoders (VAEs)
Offline reinforcement learning

Methodologically, we focus on statistical learning theory and generalization analysis of machine learning algorithms using tools such as PAC-Bayesian theory and information theory.

For examples of student projects, please refer to the list of Master’s theses and seminar textbooks used in our reading groups.

Although we are not currently working on them, other topics I have previously published on include:

Large-scale Bayesian inference and ensemble learning using variational Bayesian methods and Monte Carlo techniques
Theoretical analysis and acceleration of sampling methods (e.g., MCMC)
Development of robust learning algorithms that are less sensitive to corrupted data (e.g., outliers)
Time-series analysis using Gaussian processes

Page updated

Google Sites

Report abuse