전현호 연구실

Community detection involves dividing a network into communities where nodes within the same group are densely connected, while connections between different groups are sparse. Modularity is a key metric used to quantify the quality of these divisions.

Generative model that can generate new data instances. Generative models such as GANs, VAEs, Flow-based models and Diffusion models have been developed. Applications include image generation and various useful applications.

Link Prediction is a task in graph and network analysis where the goal is to predict missing or future connections between nodes in a network. Using graph neural networks (GNNs) to solve link prediction problems.

Inverse problems in score-based and diffusion models involve reconstructing original data from noisy, corrupted, or incomplete observations. Applications include conditional generation, image restoration, inpainting, super resolution, and medical imaging.

Single-cell RNA sequencing is used to analyze gene expression data of individual cells. Variational autoencoder has been adopted for analysis of single-cell data. Interpretable nonnegative matrix factorization method for dimension reduction.

Characterize complex dependence structure among correlated biological variables using data integration. Conditional Gaussian graphical model (CGGM) framework for modeling complex biological networks of gene-gene and gene-genome regulations.