[終了しました]ipi seminar [オンライン開催] 2022年11月10日(木)10：00～11：30
知の物理学研究センター / Institute for Physics of Intelligence (iπ)
橋本 幸士 Koji HASHIMOTO / 京都大学 Kyoto University
"Deep learning and emergent spacetime"
Formulating quantum gravity is one of the final goals of fundamental physics. Recent progress in string theory brought a concrete formulation called AdS/CFT correspondence, in which a gravitational spacetime emerges from lower-dimensional non gravitational-quantum systems, but we still lack understanding how the correspondence works.
At this end, I discuss similarities between quantum gravity and deep learning architecture, by regarding the neural network as a discretized spacetime. The application of machine learning in physics has been successful in some subjects, which I will review, and I use it to implement the AdS/CFT framework into a deep learning architecture, and show the emergence of a curved spacetime as a neural network, from a given teacher data of quantum systems.