Hello, I'm Tae-Geun Kim-Morgen 👋
- Key Laboratory of Nuclear Physics and Ion-beam Application (MOE), Fudan University & RIKEN iTHEMS
- Publications under Tae-Geun Kim
- Curriculum Vitae · Google Scholar · ORCID
- Blog
- AI for Science — neural operators, physics-informed deep learning, Hamiltonian learning
- Dark matter physics — axion-like particles, primordial black holes, detectability studies
- Scientific & High Performance Computing — numerical algorithms, Rust-based tools, parallel computing
- Comprehensive Rust numeric library for linear algebra, numerical analysis, statistics, and machine learning
- Supports automatic differentiation, special functions, DataFrame, and BLAS/LAPACK integration
- User-friendly syntax inspired by R, NumPy, and MATLAB
- Novel learning rate schedulers addressing the learning curve decoupling problem in deep learning
- Epoch-insensitive design enables stable training across varying durations without retuning
- Evaluated on image classification, time series prediction, and operator learning tasks with PyTorch
- Personal research assistant for arXiv — discover, organize, and annotate papers from the terminal
- TF-IDF content similarity + category/keyword/recency scoring for personalized recommendations
- Full TUI, AI summaries (Gemini/Claude/OpenAI/Ollama), reading lists, and export to Markdown/JSON/CSV
- Pure Rust special functions library (gamma, beta, error functions) with zero dependencies
- Lightweight implementation based on "Numerical Recipes" algorithms
- Reinforcement Learning library in Rust with modular agents, environments, and policies
- Includes Epsilon Greedy, Value Iteration, and Q-Learning implementations
More projects
- Flexible PyTorch experiment template with YAML-based configuration
- Supports multiple seeds, device selection, and LR scheduling for reproducible ML research
- Rust automatic differentiation library using computational graphs
- Forward/backward propagation with cached and non-cached gradient options
- Deep learning network for mass and momentum estimation in high-energy collider events
- Robust mass peak recovery under combinatoric uncertainties and detector smearing
-
Tae-Geun Kim, Giorgio Comitini, Jonas Grage, Benjamin Joens, Marc Schreiber, Soumya Sen, Russell R P Senthamarai and Johanna Sörngård, Peroxide: A Batteries-Included Numerical Computing Library for Rust, J. Open Source Softw. 11, 10366 (2026)
-
Yongsoo Jho, Tae-Geun Kim, Jong-Chul Park, Seong Chan Park and Yeji Park, Primordial Black Holes as a Factory of Axions: Extragalactic Photons from Axions, Prog. Theor. Exp. Phys. 2026, 023B07, arXiv:2212.11977 (2026)
-
Taehyeun Kim, Tae-Geun Kim, Anouck Girard, Ilya Kolmanovsky, Learning Unknown Gravitational Fields for Spacecraft State Estimation: Temporal Hamiltonian Neural Network-based Adaptive Kalman Filtering, under review at IEEE Trans. Aerosp. Electron. Syst., earlier version arXiv:2501.18808 (2025)
-
Tae-Geun Kim, Seong Chan Park, Neural Hamilton: Can A.I. Understand Hamiltonian Mechanics?, arXiv:2410.20951 (2024)
-
Tae-Geun Kim, HyperbolicLR: Epoch insensitive learning rate scheduler, arXiv:2407.15200 (2024)
-
Chang Min Hyun, Tae-Geun Kim, and Kyounghun Lee, Unsupervised sequence-to-sequence learning for automatic signal quality assessment in multi-channel electrical impedance-based hemodynamic monitoring, Comput. Methods Programs Biomed. 247, 108079, arXiv:2305.09368 (2024)
-
Kayoung Ban, Dong Woo Kang, Tae-Geun Kim, Seong Chan Park and Yeji Park, DeeLeMa : Missing information search with Deep Learning for Mass estimation, Phys. Rev. Research 5, 043186, arXiv:2212.12836 (2023)
- Primary Languages : Rust, Python, C++, Julia
- Frameworks & Libraries
- Numerical Computing: peroxide, numpy, scipy, pandas/polars, BLAS/LAPACK, eigen, mathematica
- Machine Learning: PyTorch, JAX/Equinox/Optax, W&B, Optuna, Candle, TensorFlow, Scikit-Learn
- Visualization: matplotlib, vegas, ggplot2, plotly
- High Energy Physics: BlackHawk, GALPROP, MadGraph, ROOT
- Quantum Computing: PennyLane, Qiskit, Cirq, RustQIP
- Web: Django, Vue, Firebase, Hugo, Zola, Elm
Github contributions