Publications

Select Publications (see Google Scholar for an exhaustive list)

    • A Level Proximal Point Method for nonconvex sparse constrained optimization. Digvijay Boob, Qi Deng, Guanghui Lan and Yilin Wang. Accepted for poster presentation at NeurIPS 2020.

    • Stochastic First-order Methods for Convex and Nonconvex Functional Constrained Optimization. Digvijay Boob, Qi Deng and Guanghui Lan. Submitted. (arxiv) Student Best Paper Award second prize at Informs Optimization Society, 2020.

    • Differentially Private Mixed-type Synthetic Data Generation for Unsupervised Learning. Uthaipon Tantipongpipat, Chris Waites, Digvijay Boob, Amaresh Ankit Siva, Rachel Cummings. Submitted. (arxiv)

    • Flowless: Extracting Densest Subgraphs Without Flow Computations. Digvijay Boob, Yu Gao, Richard Peng, Saurabh Sawlani, Charalamos Tsourakakis, Di Wang, Junxing Wang. Upadted version of this paper is accepted for oral presentation at WebConf (WWW) 2020. (arxiv)

    • Faster Width-dependent Algorithm for Mixed Packing and Covering LPs. Digvijay Boob, Saurabh Sawlani and Di Wang. Accepted for oral presentation at NeurIPS 2019. (arxiv)

    • Complexity of Training ReLU Neural Network. Digvijay Boob, Santanu S. Dey and Guanghui Lan. Accepted at Discrete Optimization. (arxiv)

Talks

    • Upcoming:

      • Level Contrained Proximal Point (LCPP) method for nonconvex sparse constrained optimization (Virtual poster at NeurIPS 2020; Mathematical and Computational Engineering, Universidad Católica de Chile).

      • Stochastic first-order methods for nonconvex nonsmooth function constrained optimization (Informs annual virtual meeting, 2020).

    • Algorithms for nonlinear optimization for functions constraints (IEOR, Columbia University; Mathematical Sciences, UT Dallas; EMIS, SMU) Jan 2020.

    • Faster width-dependent algorithm for mixed packing and covering LPs (Neural Information and Processing Systems, Vancouver, Canada) Dec 2019; An extended version was presented at ACO student seminar, Nov 2019.

    • Stochastic first-order methods for convex function constrained optimization (Informs annual meeting, Seattle) Oct 2019.

    • Complxity of trainging ReLU neural network (Informs annual meeting, Phoenix) Oct 2018.

    • Differentially private synthetic data generation using GANs (Won first prize in NIST competition, presented at TPDP 2018)

    • Theoretical properties of first order optimizer for One hidden layer Neural Network (INFORMS Annual Meeting, Oct 2017)