Peter Hyunseok Jang
Product Engineer @ SurvalentResearch @ UofT AIPS Lab
Undergraduate Student (Honours) @
University of Toronto
hyunseok [dot] jang [at] mail [dot] utoronto [dot] ca
CV (Upon request)  /  LinkedIn  /  GitHub  /  Google Scholar  /  X

About me
I am a senior undergraduate student at the University of Toronto. Currently, I am working on ML models for electric power distribution systems at Survalent, while doing research on LLMs at the AIPS Lab under the supervision of Dr. Kristen Menou.
Previously, I worked on applying ML techniques to the South Korean grid funded by the Ministry of Trade, Industry and Energy.
R&D Interests
I stand at the intersection of theoretical and applied research. I'm passionate about designing engines and intelligent agents that can operate and make decisions within cyber-physical systems, such as the smart grid. My research leverages machine learning techniques to achieve this, focusing on:
- Time-series analysis: Enabling agents to learn from and predict sensor data for informed actions and analytics.
- Multimodal learning: Equipping agents to understand and react to various real-time data streams (e.g., sensor readings, camera feeds).
- Retrieval-Augmented Generation (RAG): Allowing agents to access and utilize external knowledge for more robust decision-making.
- Graph neural networks: Helping agents reason about complex relationships within cyber-physical networks for navigation and task coordination.
By applying these techniques, I aim to develop autonomous agents capable of sensor fusion, energy management, and efficient control within complex systems.
Publications
Novel Single Group-Based Indirect Customer Baseline Load Calculation Method for Residential Demand Response
HyunYong Lee, Hyunseok Jang, Seung-Hun Oh, Nac-Woo Kim, Seong-cheol Kim, Byung-Tak Lee
IEEE Access 2021
[Paper]
[Press (Translated)]