Praneeth Vepakomma

Praneeth Vepakomma is currently a PhD student and his focus is on developing algorithms to support distributed and collaborative machine learning. He was previously a scientist at Apple, Amazon, Motorola Solutions, PublicEngines and at various startups all of which were eventually acquired.  Some of his recent works include: 

News: I have been selected as a finalist for the Facebook Fellowship, 2021 (< 3.5% out of over 2000 students worldwide).

Publications:

1. Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity (PDF)
2. AirMixML: Over-the-Air Data Mixup for Inherently Privacy-Preserving Edge Machine Learning, (PDF)
3. Differentially private supervised manifold learning with applications like private image retrieval (PDF)
4. DAMS: Meta-estimation of private sketch data structures for differentially private COVID-19 contact tracing, (PRIML and PPML joint edition, NeurIPS-2020) (PDF)
   
5. Supervised Dimensionality Reduction via Distance Correlation Maximization, Electronic Journal of Statistics, volume 12 No.1, Pages 960--984, The Institute of Mathematical Statistics and the Bernoulli Society, 2018 (Journal) (PDF)
6. Fedml: A research library and benchmark for federated machine learning (Baidu Best Paper Award at NeurIPS-SpicyFL 2020)  (PDF)
7. DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for deep neural networks (PDF), (Accepted at CVPR 2021)
8. Advances and open problems in federated learning (with, 58 authors from 25 institutions!) (PDF), (To appear in Foundations and Trends in Machine Learning -FnTML, 2020)
   
9. Splintering with distributions: A stochastic decoy scheme for private computation
10. Assessing Disease Exposure Risk With Location Histories And Protecting Privacy: A Cryptographic Approach In Response To A Global Pandemic
11. Split learning for health: Distributed deep learning without sharing raw patient data, (PDF) (Project page) (ICLR Workshop)
12. Combinatorics of Distance Covariance: Inclusion-Minimal Maximizers of Quasi-Concave Set Functions for Diverse Variable Selection,  (Discrete Applied Mathematics, Journal)  (PDF)
13. PPContactTracing: A privacy-preserving contact tracing protocol for covid-19 pandemic
14. No Peek: A Survey of private distributed deep learning (PDF)
15. Reducing leakage in distributed deep learning for sensitive health data, Accepted to ICLR 2019 Workshop on AI for social good. 
16. Optimal bandwidth estimation for a fast manifold learning algorithm to detect circular structure in high-dimensional data (PDF)
17. A Fast Algorithm for Manifold Learning by Posing it as a Symmetric Diagonally Dominant Linear System , Applied and Computational Harmonic Analysis (PDF)
18. A-Wristocracy: Deep Learning on Wrist-worn Sensing for Recognition of User Complex Activities (PDF), (IEEE Body Sensor Networks)
19. Apps gone rogue: Maintaining personal privacy in an epidemic
20. ExpertMatcher: Automating ML Model Selection for Clients using Hidden Representations, (2019 ICLR Workshop)
21. "Scoring Practices for Remote Sensing of Land Mines", Mathematical Problems in Industry, MPI Workshop, Duke University.
22. A Review of Homomorphic Encryption Libraries for Secure Computation (PDF)

Invited Talk at  Workshop on Federated Learning and Analytics at IBM Thomas J Watson Research Center, Feb 2020 https://federated-learning.bitbucket.io/ibm2020/ 

CVPR 2019 Tutorial on “Distributed Private Machine Learning for Computer Vision: Federated Learning, Split Learning and Beyond”.

We gave a half-day tutorial at CVPR 2019: On Distributed Private Machine Learning for Computer Vision: Federated Learning, Split Learning and Beyond 

Organizing DPML Workshop at ICLR,

Organized Workshop on Split Learning for Distributed Machine Learning (SLDML’21), https://splitlearning.github.io/workshop.html