Working Experiences

  • Quantitative Researcher

          Citadel Securities

          Sep. 2019 - Present

          New York, NY

  • Quantitative Research Intern

          Citadel Securities

          June 2018 - Sep. 2018

          Chicago, IL

  • Research-based Software Engineer with Le Song

          Ant Financial

          Dec. 2017 - Jan. 2018

          Hangzhou, China

          UC Berkeley

          Fall 2017

          Berkeley, CA

          Microsoft Research

          Summer 2017

          Redmond, WA

Publications

. HopSkipJumpAttack: A Query-Efficient Decision-Based Attack. In IEEE S&P (20-min Talk), 2020.

Preprint Code Project Video

. ML-LOO: Detecting Adversarial Examples with Feature Attribution. In AAAI (Spotlight), 2020.

Preprint Code Project Poster Slides

. Greedy Attack and Gumbel Attack: Generating Adversarial Examples for Discrete Data. In Journal of Machine Learning Research, 2020.

Preprint Project

. L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Data. In ICLR (Poster), 2019.

Preprint Code Project Poster

. Learning to Explain: An Information-Theoretic Perspective on Model Interpretation. In ICML (20-min Oral), 2018.

Preprint Code Project Oral Poster

. Language-Based Image Editing with Recurrent Attentive Models. In CVPR (Spotlight), 2018.

PDF Code Project Spotlight Talk Poster

. DAGGER: A sequential algorithm for FDR control on DAGs. Biometrika, 2018.

Preprint Code Project Slides

. Kernel Feature Selection via Conditional Covariance Minimization. In NIPS, 2017.

PDF Code Project Poster Blog

. Non-Convex Finite-Sum Optimization Via SCSG Methods. In NIPS, 2017.

PDF Code Project Poster

. QuTE: Decentralized multiple testing on sensor networks with false discovery rate control. In CDC, 2017.

PDF Code Project Poster

Projects

Model Interpretation and Feature Selection

We focus on feature selection and instancewise feature selection as approaches for model interpretation.

Adversarial Robustness

We design systematic methods to evaluate and improve the robustness of machine learning models.

Decision Making on Graphs

We design algorithms for hypothesis testing on graphs and networks that control false discovery rate.

Language-based Image Editing

We design models for editing images based on language description.

Optimization

We study variance reduction in stochastic gradient optimization.

Professional Activities

  • Talk on HopSkipJumpAttack: A Query-Efficient Decision-Based Attack.
    IEEE Symposium on Security and Privacy, May 2020.

  • Oral and Poster on LS-Tree: Model Interpretation When the Data Are Linguistic.
    AAAI, Feb 2020.

  • Spotlight Talk and Poster on ML-LOO: Detecting Adversarial Examples with Feature Attribution.
    AAAI, Feb 2020.

  • Talk on model interpretation and adversarial robustness.
    Shanghai Jiao Tong University, July 2019.

  • Talk on model interpretation and adversarial robustness.
    Facebook, June 2019.

  • Poster on L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Data.
    International Conference on Learning Representations (ICLR), May 2019.

  • Oral and Poster on Learning to Explain: An Information-Theoretic Perspective on Model Interpretation.
    International Conference on Machine Learning (ICML), July 2018.

  • Spotlight Talk and Poster on Language-based Image Editing with Recurrent Attentive Models.
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2018.

  • Poster on Kernel Feature Selection via Conditional Covariance Minimization.
    31st Neural Information Processing Systems (NIPS), December 2017.

  • Poster on Nonconvex Finite-Sum Optimization Via SCSG Methods. 31st Neural Information Processing Systems (NIPS), December 2017.

  • Talk on FDR control on directed acyclic graphs. 10th International Conference on Multiple Comparison Procedures, June 2017.

  • Poster on Decentralized decision making on networks with FDR control. 10th International Conference on Multiple Comparison Procedures, June 2017.

Honors & Awards

Contact

  • jianbochen at berkeley dot edu