Embedded Files
Welcome
Hello, my name is Misaki Ozawa. I am a CNRS researcher at Laboratory for Interdisciplinary Physics (LIPhy) of Univ. Grenoble Alpes (UGA). Additionally, I hold a junior chair position at Multidisciplinary Institute of Artificial Intelligence (MIAI). My research interests lie in the fields of statistical physics, disordered systems, and machine learning.
Research
Research
I am interested in understanding the nature of glass and the phenomenon of the glass transition. We use glass every day ? it's an essential part of our daily lives. Actually, glass is one of the oldest materials produced by humanity, with its origins dating back to about 5000 years ago, even before the era of iron! When you look at the microscopic structure of glass, it appears liquid-like. Yet, at a macroscopic level, as we experience daily, glass behaves like a solid. Despite over 70 years of intensive research, we still lack a clear understanding of its true nature. Is glass simply a liquid with such high viscosity that it can effectively be considered a solid? Or does glass reflect a unique thermodynamic phase? This question presents a big challenge in modern physics.
I am interested in understanding the nature of glass and the phenomenon of the glass transition. We use glass every day ? it's an essential part of our daily lives. Actually, glass is one of the oldest materials produced by humanity, with its origins dating back to about 5000 years ago, even before the era of iron! When you look at the microscopic structure of glass, it appears liquid-like. Yet, at a macroscopic level, as we experience daily, glass behaves like a solid. Despite over 70 years of intensive research, we still lack a clear understanding of its true nature. Is glass simply a liquid with such high viscosity that it can effectively be considered a solid? Or does glass reflect a unique thermodynamic phase? This question presents a big challenge in modern physics.
I am also interested in interdisciplinary studies between physics and machine learning. Nowadays, machine learning is becoming a powerful tool for physicists to analyze various physics problems. Conversely, the concepts and techniques developed in physics help us uncover the mystery behind the outstanding performance of machine learning architectures (such as deep neural networks), which have been considered as black boxes. Currently, we are developing machine learning methods inspired by statistical physics, making synergy between physics and machine learning.
I am also interested in interdisciplinary studies between physics and machine learning. Nowadays, machine learning is becoming a powerful tool for physicists to analyze various physics problems. Conversely, the concepts and techniques developed in physics help us uncover the mystery behind the outstanding performance of machine learning architectures (such as deep neural networks), which have been considered as black boxes. Currently, we are developing machine learning methods inspired by statistical physics, making synergy between physics and machine learning.
Publication
Current Team Members
Bruno Ventéjou (Postdoc)
Takumi Nagasawa (PhD)
Alexandre Stepanetz (Master)
Ege Adiguzel (Undergrad)
Former Team Members
Nina Javerzat （CNRS, Institut Fourier）
Gerhard Jung (University of Innsbruck)
Manar Shehazi (Mines Paris - PSL)
Bahaa Mazloum (European Synchrotron Radiation Facility)
Anand Sharma (Indian Institute of Science Education and Research)
Samuel Kubiti (University of Cape Coast)
Contact
Address: Laboratoire Interdisciplinaire de Physique (LIPhy), Universite Grenoble Alpes, 140 Rue de la Physique, 38402 Saint-Martin-d'Heres, France
Mail Address: misaki.ozawa@univ-grenoble-alpes.fr
Page updated
Google Sites
Report abuse