Steven Liu Kin Sum (廖健森)

I am a PhD candidate at Computer Science Department of Stony Brook University. Resume is available HERE.

My adviser is Prof. Jie GaoI also collaborate with Prof. Shan Lin from ECE Department. I am grateful to do internships at FiveAI, Yahoo Research and Bosch RTC previously to enrich my research capability.

Recent working projects include Graph Classification, Graph Embedding Learning and Privacy Mechanism in Machine Learning models.

Research Interests:
  • Machine Learning: Representation Learning, Deep Learning, Adversarial Learning
  • Control Theory: Human-in-the-loop control for Cyber Physical Systems, Deep Reinforcement Learning
  • Combinatoric and Geometric Algorithms: Wireless Sensor Networks (routing, scheduling, coverage)
Drafts:
  • Performing Co-Membership Attacks Against Deep Generative Models
    [preprint]
  • Knowledge Graph Representation Learning for Entity Recommendation
    [soon]
  • Topology based scalable graph kernels using Ricci Curvature
    [soon]
Publications:
  • Patrol Scheduling Against Adversaries with Varying Attack Durations (AAMAS 2019)
    [soon]
  • Optimizing Sensor Deployment with Line-Of-Sight Constraints: Theory and Practice (EWSN 2019)
    [soon]
  • On-Street Parking Guidance with Real-Time Sensing Data for Smart Cities (SECON 2018)
    [soon]
  • Understanding Occupancy Patterns in a Commercial Space (TC-CPS 2018)
    [soon]
  • Data-Driven Robust Control for Type-1 Diabetes Under Meal and Exercise Uncertainties (CMSB 2017)
    [link][slide][pdf] (Slide credit to Nicola Paoletti)
  • Reliable Stream Scheduling with Minimum Latency for Wireless Sensor Networks (SECON 2017)
    [link][pdf]
  • Joint Sensing Duty Cycle Scheduling for Heterogeneous Coverage Guarantee (INFOCOM 2017)
    [link][slide][pdf]
  • Joint Sensor Duty Cycle Scheduling with Coverage Guarantee (MobiHoc 2016)
    [link][slide][pdf]
  • Efficient Multi-Channel Communications in Wireless Sensor Networks (TOSN 2016)
    [link][pdf]
  • Probabilistic Convex Hull Queries over Uncertain Data (TKDE 2014)
    [link][pdf]
  • Finding Distance-Preserving Subgraphs in Large Road Networks (ICDE 2013)
    [link][pdf]
Posters:
  • COD: A Dataset of Commercial Building Occupancy Traces (BuildSys 2017)
    [link]
  • Long Term Occupancy Estimation in a Commercial Space: An Empirical Study (IPSN 2017)
    [link]
  • Combinatorics, algorithms and systems for sensor deployment with line-of-sight constraints (MobiHoc 2016)
    [link]
  • Joint Sensor Duty Cycle Scheduling (FWCG 2016)
    [slide][pdf]