Received Sylff fellowship in 2023
Academic supervisor: Dr. Jin Nakazawa
Current affiliation: Graduate School of Media and Governance, Keio University
Hirokono Kawashima is a Ph.D. student at Keio University, specializing in continual learning of deep learning models and their applications. Continual learning is a field that aims to enable machine learning models to learn sequential tasks as flexibly as humans do, without forgetting what they have learned in the past. Currently, she is working on an improved replay method for realizing continual learning, in which a portion of the data from previous tasks is stored in a memory buffer as "memory" and used for new learning.
Nakazawa Lab. (https://www.jn.sfc.keio.ac.jp/), where she belongs, has been conducting a number of joint research projects with local governments and municipalities on smart cities. Her goal is to build flexible deep learning models that can adapt to changes in the real world and to link the design of learning methods to real-world applications of deep learning technology.