Skip to content
View bangyen's full-sized avatar
9 followers · 2 following

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLO

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLO

Block or report bangyen

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
bangyen/README.md

Hey, I’m Bangyen

I build systems that blend AI, distributed computing, and optimization.

Featured Projects

  • ZSharp: +5.26% accuracy and 4.4× faster training with Sharpness-Aware Minimization, Apple Silicon optimized.
  • Oligopoly: 98.5% collusion detection, 1e-6 precision, and 72.3% strategy adaptation in advanced economic simulations.

Publications

  • Generalized Collective Algorithms for the Exascale Era — CLUSTER 2023
  • Revisiting Computation for Research: Practices and Trends — SC 2024

Quick Links

Pinned Loading

  1. wikipedia wikipedia Public

    AI-powered Wikipedia article maturity scoring with feature extraction, LightGBM calibration, and FastAPI + CLI interfaces.

    Python

  2. chess chess Public

    Explainable chess engine with 86.7% decisive faithfulness, 2.5 sparsity explanations, and 100% position coverage via Rust-accelerated Stockfish probes.

    Python

  3. zsharp zsharp Public

    Sharpness-Aware Minimization with Z-Score gradient filtering achieving +5.26% accuracy, Apple Silicon optimized (4.39x speedup), and fully reproducible.

    Python 1

  4. oligopoly oligopoly Public

    Advanced oligopoly market simulation with validated collusion detection, research-grade precision (1e-6), and adaptive strategy learning (Q-learning).

    Python 1

  5. emergent emergent Public

    Multi-agent emergent language learning framework for studying communication protocols in referential games using Gumbel-Softmax and neural agents.

    Python

  6. regulator regulator Public

    Advanced market collusion detection achieving 93% accuracy using ML + LLMs, with real-time monitoring dashboards and validated economic modeling.

    Python