Quantum BC Seminar Series – Jinyang Li from RIKEN

Quantum BC Seminar Series Tuesday, February 10, 2026 at 2:00pm BC Time with Jinyang Li

Join on Zoom: https://ubc.zoom.us/j/69443327772?pwd=TGhhTXFIQ3ZiUmNrN0pUa3FObTNydz09

Meeting ID: 694 4332 7772 Passcode: 996727

Seminar Title: Neural Operator Learning in Quantum Simulations

Seminar Abstract:

Simulating the real-time operator dynamics is crucial to characterize quantum many-body systems. The simulation can be performed efficiently using quantum computers and potentially exhibiting a practical quantum advantage over classical computers. However, due to decoherence in quantum simulators, the operator dynamics is limited within a short evolution time, such that the quantum systems are characterized with low accuracy. On the other hand, given the short-time data, neural networks with physics priors can potentially learn the operator dynamics and predict the long-time behavior. In this work, we implement neural ordinary differential equations (Neural ODEs) to learn the intrinsic dynamics from the short-time signal and extract physical kernels. By exploiting physical symmetries and locality, it reduces complexity while retaining essential operator dynamics. Our method provides a scalable and data-efficient avenue to characterize quantum many-body systems using noisy quantum computers where long-time dynamics is unavailable.

Short Bio:

Junior Research Associate, Division of Fundamental Mathematical Science, RIKEN Center for Interdisciplinary Theoretical and Mathematical Sciences (iTHEMS)

Student Trainee, Division of Fundamental Mathematical Science, RIKEN Center for Interdisciplinary Theoretical and Mathematical Sciences (iTHEMS) (concurrent)

My research at ITHEMS aligns with Quantum computing and machine learning. I have experience in optical quantum sensors and QCD thermal history.

Recently, I have become interested in utilizing both quantum and computer techniques to figure out some hints of underlying physics rules that are hard to discover directly. I am also interested in how closely a ML model can imitate a human, or more precisely, a physicist. For example, using ML method to study the intrinsic dynamics of a quantum process and make a long-term prediction.