[Tensor Learning Team Seminar] - Talk by Dr. Rong-Yang Sun (The Chinese University of Hong Kong) “Autonomous Generation of High-Performance Tensor Network Applications with Agentic AI”

Date and time: Thursday, November 19, 2026, 14:00–15:00 JST
Format: Hybrid
Language: English

Online participation: Open to everyone via Zoom. Please register to receive the connection details.

In-person attendance: Limited to RIKEN members.
Venue: Open Space, RIKEN AIP Nihonbashi Office

Speaker: Dr. Rong-Yang Sun, Research Assistant Professor, The Chinese University of Hong Kong

Schedule:
14:00–14:45 Talk
14:45–15:00 Q&A

Abstract:

Tensor network methods are widely used in quantum physics and machine learning for efficiently representing high-dimensional objects, including quantum many-body wave functions, complex data, and neural network layers. However, developing high-performance tensor network software for large-scale applications remains challenging. Recent advances in large language model (LLM)-driven agentic AI are transforming software development, yet autonomous programming for high-performance computing remains difficult.

In this talk, I will present a fully autonomous agentic workflow for generating and optimizing high-performance tensor network applications. I will demonstrate how this approach can achieve performance comparable to state-of-the-art implementations developed by human experts when solving challenging realistic computational problems. I will further introduce the Tensor Computing API (TCAPI), a unified, cross-platform interface for tensor computations, and explain how its clear abstraction boundaries and concise interface contracts play a crucial role in enabling autonomous development of high-performance tensor network applications.

Bio:

Rong-Yang Sun received his PhD in Physics from Tsinghua University in 2021. From 2021 to 2025, he worked at RIKEN as a Postdoctoral Researcher and Research Scientist. He then worked briefly at California State University Northridge before joining the Chinese University of Hong Kong as a Research Assistant Professor in 2026. His research interests include computational quantum many-body physics, quantum computing, tensor network methods, and high-performance computing.