President and CEO 
 Tohru Asami 

This year, ATR marks the 40th anniversary of its founding. Last year, Keihanna Expo 2025 was held in conjunction with Expo 2025 Osaka, Kansai, Japan, under the themes of “Co-Creation, Demonstration, and Commercialization.” In a symposium held as part of this event, titled “Learning from the Past to Create the Future,” I was once again given the opportunity to reflect upon the achievements of our predecessors.

In 1988, ATR initiated research on speech recognition using neural networks (NNs), an effort broadly led by Waibel, Hanazawa, Hinton, Shikano, and Lang. Dr. Ken-ichi Funahashi provided a mathematical foundation for this work through the universal approximation theorem, which demonstrated that arbitrary continuous mappings could be approximated by neural networks with three or more layers.

Today, no one would deny the scientific significance of this theoretical foundation of deep learning. At the same time, it is also true that we ourselves have not always given this great achievement the recognition it deserves. As President of ATR, I felt a responsibility to seriously consider this apparent neglect. Thus, I concluded that ATR exists today thanks to the efforts of those predecessors who, in ways both celebrated and unsung, have devoted themselves to supporting research and society.

After a long struggle toward practical commercialization, information science is now entering a transformative era of symbiosis with avatars and AI. Yet we still do not fully understand the fundamental matter of what “information” truly is. Beyond implementing current technologies in society, I believe we also have the important mission of continuously pursuing research that explores the underlying essence of emerging technologies. I feel that 2025 marked an important turning point, when society as a whole began taking new steps in that direction.

We warmly appreciate your continued support and encouragement.