Cong Han Lim

ML & Optimization Researcher

last updated: 02/15/2023

I am a Member of Research Staff at Voleon. I obtained my PhD in Computer Science at UW-Madison working with Jeff Linderoth, Jim Luedtke, and Steve Wright. Then spent some time as a postdoc at Wisconsin Institute for Discovery, Georgia Tech ISyE, and Uber ATG.

My academic research is in the field of Mathematical Optimization. I developed algorithms for machine learning and operations research, generally with the goal of making hard problems more tractable. Some topics I’ve tackled include robustness verification, structured regression, parallel optimization, and mixed-integer nonconvex optimization.

Papers

  • Hierarchical Verification for Robustness Verification
    Cong Han Lim, Raquel Urtasun, Ersin Yumer
    ICML 2020
  • Subgradient Sampling Methods for the Lagrangian Dual in Stochastic Mixed-Integer Programming
    Cong Han Lim, Jeffrey Linderoth, James Luedtke, Stephen Wright
    INFORMS Journal on Optimization (2020)
  • An Efficient Pruning Algorithm for Robust Isotonic Regression
    Cong Han Lim
    NeurIPS 2018
  • A Distributed Quasi-Newton Algorithm for Primal and Dual Regularized Empirical Risk Minimization
    Ching-Pei Lee, Cong Han Lim, Stephen Wright
    ArXiv (2019) / KDD 2018 (shorter version)
  • A Note on Extended Formulations for Cardinality-based Sparsity
    Cong Han Lim
    NeurIPS Optimization for Machine Learning Workshop 2017
  • k-Support and Ordered Weighted Sparsity for Overlapping Groups - Hardness and Algorithms
    Cong Han Lim, Stephen Wright
    NeurIPS 2017
  • Valid Inequalities for Separable Concave Constraints with Indicator Variables
    Cong Han Lim, Jeff Linderoth, James Luedtke
    Mathematical Programming (2017) / IPCO 2016
  • A Box-Constrained Approach for Hard Permutation Problems
    Cong Han Lim, Stephen Wright
    ICML 2016
  • Efficient Bregman Projections onto the Permutahedron and Related Polytopes
    Cong Han Lim, Stephen Wright
    AISTATS 2016
  • Sorting Network Relaxations for Vector Permutation Problems
    Cong Han Lim, Stephen Wright
    NeurIPS 2014
  • Validating Sample Average Approximation Solutions with Negatively Dependent Batches
    Jiajie Chen, Cong Han Lim, Peter Qian, Jeffrey Linderoth, Stephen Wright
    ArXiV (2014)

    • Other Stuff

    A good chunk of my free time geeking out over films. In Chicago, check out Doc Films, Facets Cinematheque, Gene Siskel Film Center, and Music Box Theatre. In Madison, the UW Cinematheque has impeccable programming. Madison is also home to film theorists David Bordwell and Kristin Thompson, who have one of my favorite film blogs. In Los Angeles, check out everything. The Academy Museum is not a bad place to start. If you want streaming sites that offer something a little different, try Mubi or The Criterion Channel.

    last movie watched: The Birds

    I am from Singapore. We have great food there.