Research Portfolio

  1. Crafting Adversarial Examples for Neural Networks
    • ZOO: powerful black-box attack to neural networks
    • EAD: crafting L1 norm based adversarial examples - better attack transferability!
  2. Community Detection: Theory and Algorithms
    • Phase transition analysis of community detection under general connectivity models
    • AMOS & MIMOSA: theory-driven automated community detection algorithms for single-layer and multi-layer graphs 
    • Deep (core) community detection 
    • SGC-GEN: pseudo-supervised community detection meta algorithm
  3. Network Analytics
    • Structural feature extraction from a single graph or a graph sequence
    • Anomaly detection based on graph connectivity
  4. Optimization
    • Accelerated distributed dual averaging over networked agents
    • Zeroth-order ADMM: convergence and algorithm
  5. Event Propagation and Control in Networks
    • Modeling malware propagation in heterogeneous networks
    • Event propagation control via node and edge patching
  6. Network Resilience
    • LFVC: effective centrality attack for network disruption 
    • Sequential and game-theoretic information fusion for defending connectivity attacks
(Last updated in Oct. 2017)