Skip to content
View lucaswychan's full-sized avatar

Highlights

  • Pro

Block or report lucaswychan

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, 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
lucaswychan/README.md

Portfolio LinkedIn X PyPI

About Me

I'm Lucas Chan, an MPhil candidate at HKUST supervised by Prof. Yangqiu Song. My research spans embedding model and LLM reasoning. I hold a Bachelor's degree in Computer Science & Electronic Engineering (First Class Honours) from HKUST.

Projects

  • Neural Network Framework (C++): A PyTorch-like framework built from scratch in C++ (STL only), where modules can be dynamically stacked without considering the implementation of forward and backward propagation. Currently support MLP and CNN.
  • Neural Stock Prophet: A PyPI package for stock forecasting using LSTM, Attention, and ARIMA.
  • Federated Edge AI for 6G: Addressed computation and communication heterogeneity in federated learning using FedDyn and RIS-FL.

Current Interests

  • Embedding model
  • LLM reasoning
  • Reinforcement learning
  • Model architectures

Hobbies

  • Coding (daily LeetCode)
  • AI News & Trends
  • Stock Trading
  • Gym

Pinned Loading

  1. neural-stock-prophet neural-stock-prophet Public

    LSTM-ARIMA with attention mechanism and multiplicative decomposition for sophisticated stock forecasting.

    Python 4 1

  2. Federated-Edge-AI-For-6G Federated-Edge-AI-For-6G Public

    Fusion of federated leaerning algorithm and reconfigurable intelligent surface from 6G

    Python 6 1

  3. neuralnet-cpp neuralnet-cpp Public

    A PyTorch-like neural network framework in pure C++ from scratch, using only C++ STL.

    C++ 2

  4. quant-lvlm quant-lvlm Public

    Easy-to-use large vision language model pipeline for quantitative analysis

    Python 3

  5. Retinal-Vessel-Segmentation Retinal-Vessel-Segmentation Public

    Use Autoencoder to perform automatic segmentation of retinal vessels.

    Python 2

  6. contrastive-resnet50 contrastive-resnet50 Public

    Employ contrastive learning to enhance the ResNet-50 performace for skin lesion classification.

    Python 3