Palma London

Ph.D. in Computer Science at Caltech
plondon (at) caltech (dot) edu

.
Biography

I am a Postdoctoral Researcher at UCSD in the Halicioglu Data Science Institute and the TILOS NSF AI Institute.

I received my Ph.D. in Computer Science at Caltech, where I was advised by Adam Wierman.

My research focuses on mathematical optimization and distributed algorithms. I received a double B.S. degree in Electrical Engineering and Math at the University of Washington. I received both the NSF Graduate Research Fellowship (GRFP) and the Amazon Fellowship in Artificial Intelligence during my PhD. I was recently named a Rising Star in EECS.

.
Research Interests

My research broadly spans mathematical optimization and distributed algorithms. I use tools from randomized linear algebra and convex optimization theory to develop novel optimization algorithms that are amenable to distributed and parallelized computation.

I work on developing efficient algorithms for solving extremely large, prohibitively massive optimization problems. My general approach is to develop techniques to compress optimization problems, where the goal is to represent a huge problem as a smaller problem, or set of smaller sub-problems. Despite only using partial information from the original problem, we can still ensure provable guarantees on solution quality and run time. Often this approach is especially feasible if we can leverage specific structure or characteristics that may exist in the problem data or constraints. This approach is highlighted in the following projects:

A Parallelizable Acceleration Framework for Packing Linear Programs
Palma London, Shai Vardi, Adam Wierman, Hanling Yi.
Association for the Advancement of Artificial Intelligence (AAAI) 2018. pdf Poster

We accelerate (speed up) state-of-the-art commercial linear program solvers by two orders of magnitude, while maintaining a near-optimal solution.



Logarithmic Communication for Distributed Optimization in Multi-Agent Systems
Palma London, Shai Vardi, Adam Wierman.
Proceedings of the ACM on Measurement and Analysis of Computing Systems 3(3): 1-29, 2019. pdf Talk



Speeding up Linear Programming using Randomized Linear Algebra
Agniva Chowdhury, Palma London, Haim Avron, and Petros Drineas.
To appear in Advances in Neural Information Processing Systems (NeurIPS), 2020. pdf Code


.
Education

Ph.D.         Computer Science, Caltech, 2020
B.S.E.E.   Electrical Engineering, University of Washington, 2014
B.S.           Mathematics

.
Publications

Black-Box Acceleration of Monotone Convex Program Solvers
Palma London, Shai Vardi, Reza Eghbali, Adam Wierman.
Operations Research (OR), 2022. pdf

Faster Randomized Interior Point Methods for Tall/Wide Linear Programs
Agniva Chowdhury, Gregory Dexter, Palma London, Haim Avron, and Petros Drineas.
To appear in Journal of Machine Learning (JMLR), 2022. pdf

Frameworks for High Dimensional Convex Optimization
Palma London. Ph.D. Thesis, California Institute of Technology, 2020. pdf

Speeding up Linear Programming using Randomized Linear Algebra
Agniva Chowdhury, Palma London, Haim Avron, and Petros Drineas.
To appear in Advances in Neural Information Processing Systems (NeurIPS), 2020. pdf Code

Logarithmic Communication for Distributed Optimization in Multi-Agent Systems
Palma London, Shai Vardi, Adam Wierman.
Proceedings of the ACM on Measurement and Analysis of Computing Systems 3(3): 1-29, 2019. pdf Talk

A Parallelizable Acceleration Framework for Packing Linear Programs
Palma London, Shai Vardi, Adam Wierman, Hanling Yi.
Association for the Advancement of Artificial Intelligence (AAAI) 2018. pdf Poster

Datum: Managing Data Purchasing and Data Placement in a Geo-Distributed Data Market
Xiaoqi Ren, Palma London, Juba Ziani, Adam Wierman.
IEEE/ACM Transactions on Networking 26 (2), 893-905, 2018. pdf

Distributed optimization via local computation algorithms
Palma London, Niangjun Chen, Shai Vardi, Adam Wierman.
ACM SIGMETRICS Performance Evaluation Review 45(2): 30-32, 2017. pdf

Joint Data Purchasing and Data Placement in a Geo-Distributed Data Market
Xiaoqi Ren, Palma London, Juba Ziani, Adam Wierman.
ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Science, 2016. pdf

Learning Graphical Models with Hubs
Kean Ming Tan, Palma London, Karthik Mohan, Su-In Lee, Maryam Fazel, Daniela Witten.
Journal of Machine Learning Research (JMLR), 15(Oct):3297-3331, 2014. pdf

Node-Based Learning of Multiple Gaussian Graphical Models
Karthik Mohan, Palma London, Maryam Fazel, Daniela Witten, Su-In Lee
Journal of Machine Learning Research (JMLR), 15(Feb):445-488, 2014. pdf

.
Fellowships and Awards

TRIPODS Postdoctoral Fellowship, 2020 - 2021
Amazon Fellowship in Artificial Intelligence, 2018 - 2019
NSF Graduate Research Fellowship Program (GRFP), 2014 - 2017

.
Contact

Contact: plondon (at) caltech (dot) edu

eXTReMe Tracker

.