Independent Researcher

Joun Won

Interested in interpretable RL | Co-founder of Open-Age Academia

About

I am an independent researcher and a student at Haeryong High School.

My research focuses on developing XAI-driven control algorithms and interpreting reinforcement learning policies, with applications to aerospace systems — particularly orbital dynamics. I am broadly interested in making autonomous control systems that are interpretable.

Outside of research, I co-founded Open-Age Academia, after running into the same wall as a high-school researcher. Open-Age Academia exists to make that path clearer for the next person.

Explainable AI
Reinforcement Learning
Control Theory
Aerospace & Orbital Dynamics

Publications

2026

What Do RL Spacecraft Guidance Policies Learn? Symbolic Distillation against a Known Control Law

Joun Won

NeurIPS 2026 Workshop Interp4Discovery

Paper Code

BibTeX

@inproceedings{won2026what,
title={What Do {RL} Spacecraft Guidance Policies Learn? Symbolic Distillation against a Known Control Law},
author={Joun Won},
booktitle={NeurIPS 2026 Workshop on Interpretability for Discovery},
year={2026},
url={https://openreview.net/forum?id=XTuN0nml2m}
}

CV

Awards & Honors

2026

Korean Young Physicists' Tournament (KYPT) — Encouragement Award & Kwon Myeonghui Award, Team Haeryong CASS

2026

KENTECH 3rd Symposium on Sustainable Energy Transition (SSET) — Best Poster Award

Contact

I like getting mails.