Junsung (June) Hwang

Code

Scalable Construction of Spectrally Near-Optimal Networks via Reinforcement Learning

(This project has also been referred to as "Spectral Optimal Graph Construction via Reinforcement Learning" in earlier drafts.)
SeaWulf-scale reinforcement-learning pipeline that edits graph topologies to maximize algebraic connectivity, benchmarking against effective-resistance heuristics and SDP relaxations.

Skills: C, Python, PyTorch, NetworkX, CVXPY, MOSEK, Slurm, HPC, Linux

GraphEvo

Open-source evolutionary search framework with a Master Regulatory Gene crossover, MPI-based island model, and reproducible CLI/API for high-connectivity graph discovery.

Skills: C++, Python, MPI, OpenMP, GitHub Actions, Slurm, HPC, Linux, Git

A New Broadcast Model for Several Network Topologies

Equal-contribution co-authorship on BBS, a broadcast model tuned to specific network topologies that delivers over 30% speedup versus classical schedulers on SeaWulf.

Skills: Python, MPI, Parallel Profiling, Slurm, HPC, Linux, Docker, Git

CTGNN

Proved invariance of the contraction metric $P(\beta_0)$ under admissible graph edits and derived a rank-2 Laplacian update theory.

Skills: Mathematics