Vivek Bharadwaj

  • Program Year: 4
  • Academic Institution: University of California, Berkeley
  • Field of Study: HPC/Scientific Computing
  • Academic Advisor: James Demmel, Aydin Buluc
  • Practicum(s):
    National Renewable Energy Laboratory (2022)
  • Degree(s):
    B.S. Computer Science and Mathematics, California Institute of Technology, 2020
  • Personal URL: http://vivek-bharadwaj.com

Summary of Research

I study computational kernels to accelerate a broad spectrum of scientific computing applications; much of this work focuses on minimizing communication, which encompasses both DRAM-cache memory traffic in shared memory settings and processor-to-processor communication in distributed memory settings. I'm thrilled to be advised by James Demmel and Aydin Buluc at UC Berkeley.

Some of my recent interests include mixed precision matrix multiplication and various operations involving sparse and dense matrices, such as sampled-dense-dense matrix multiplication. Applications of the latter kernel include matrix factorization, document distance clustering, and graph neural network training.

I earned my undergraduate degree at Caltech in mathematics and computer science. I was fortunate enough to receive mentorship from Chris Umans, Mikhail Shapiro, and Rose Yu. You can find more details about me on my personal website.

Publications

V. Bharadwaj, O. A. Malik, R. Murray, A. Buluc, J. Demmel, "Distributed-Memory Randomized Algorithms for Sparse Tensor CP Decomposition", Proceedings of the 36th ACM Symposium on Parallelism in Algorithms and Architectures, Jun 2024.


V. Bharadwaj, O. A. Malik, R. Murray, L. Grigori, A. Buluc, J. Demmel, "Fast Exact Leverage Score Sampling from Khatri-Rao Products with Applications to Tensor Decomposition", Conference on Neural Information Processing Systems (2023).

O. A. Malik, V. Bharadwaj, R. Murray, "Sampling-Based Decomposition Algorithms for Arbitrary Tensor Networks", arXiv preprint arXiv:2210. 03828, 2022.

V. Bharadwaj, A. Buluc, J. Demmel, "Distributed Memory Sparse Kernels for Machine Learning," 36th IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2022.

P. Ramesh et al., Ultraparamagnetic Cells Formed through Intracellular Oxidation and Chelation of Paramagnetic Iron, Angew Chem Int Ed Engl, vol. 57, no. 38, pp. Sep. 2018, doi: 10.1002/anie.201805042.

Awards

Berkeley Teaching Effectiveness Award, 2024.

Berkeley Outstanding Graduate Student Instructor, 2023.

NSF Graduate Research Fellowship Program Honorable Mention, 2019-2020 Application Cycle.

Thomas A. Tisch Prize for Undergraduate Teaching: Awarded at Caltech for outstanding teaching / course development in CS38: Algorithms over a three-year period.

Ph11 Scholar, Caltech: Awarded for outstanding creative problem-solving as a freshman at Caltech.