Who/Where: PhD at Massachusetts Institute of Technology from the Camera Culture group on distributed and private computation for machine learning, statistical inference and data science at large. This includes a major focus on responsible/trustworthy AI. - (Fellowship) Won the Meta (previously FB) 2022 Phd Research Fellowship (37 fellows selected out of 2300 applicants). Featured on MIT News with three other fellowship recipients.
- (Award) Non-Profit I co-founded (Integrity Distributed) won Financial Times Digital Innovation Award. Borne into fruition based on my research work.
- Check out our exciting work on Effects of Privacy-Inducing Noise on Welfare and Influence of Referendum Systems (PDF) that bridges differential privacy with social choice theory using analysis of Boolean fns.
- (Scholarship) Selected as a SERC Scholar (Social and Ethical Responsibilities of Computing Scholar) by MIT’s Schwarzman College of Computing.
- My work on split learning featured on MIT Technology Review and MIT News.
- (Award) NoPeek-Infer won a Mukh Best Paper Runner Up Award at IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021) conference.
- (Award) FedML work won a Baidu Best Paper Award at NeurIPS 2020-SpicyFL.
- Am a mentor at the upcoming United Nations Privacy Enhancing Technologies/U.N PET lab's first global hackathon.
- (Award) Won the FL-IJCAI'22 Best Student Paper Award for "Visual Transformer Meets CutMix for Split Learning", at the International Workshop on Trustworthy Federated Learning in Conjunction with IJCAI 2022 (FL-IJCAI'22).
- Attending Boston DP Summer School
- (Award) Extra Mile award at PublicEngines (acquired by Motorola Solutions)
Selected works:
I welcome requests for high-level insight, in-depth one/two-way feedback, collaboration and invited talks (check contact me).
36. Posthoc privacy guarantees for collaborative inference with modified Propose-Test-Release, (PDF), @NeurIPS 2023 Conference, Thirty-seventh Conference on Neural Information Processing Systems, Abhishek Singh, Praneeth Vepakomma, Vivek Sharma, Ramesh Raskar -Topic: Differential privacy, formalizes privacy of informal ML pipelines, Distributed/Collabirative Inference (2023)
35. Parallel quasi-concave set function optimization for scalability even without submodularity (PDF), @IEEE High Performance Extreme Computing Conference, Praneeth Vepakomma, Yulia Kempner, Rodmy Paredes Alfaro, Ramesh Raskar -Topic: Parallel Combinatorial Optimization (2023)
34. Effects of Privacy-Inducing Noise on Welfare and Influence of Referendum Systems (PDF), Suat Evren, Praneeth Vepakomma, Topic: Differential Privacy, Social choice theory (2023)
33. Differentially private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices (PDF), @TMLR, Transactions on Machine Learning Research, Saiteja Utpala, Praneeth Vepakomma, Nina Miolane - (Journal), Topic: Geometric Statistics, Differential Privacy, Differential Geometry (2023),
32. Private mechanisms for nonlinear correlations and independence testing with energy statistics (PDF), Praneeth Vepakomma, Mohammad Mohammadi Amiri, Subha Nawer Pushpita, Clément Canonne, Ramesh Raskar, Alex Pentland - Topic: Statistics, Independence Testing, Private Correlations, Differential Privacy (2022)
31. Differentially private CutMix for Split Learning with Vision Transformer (PDF), Seungeun Oh, Sihun Baek, Hyelin Nam, Seong-Lyun Kim, Praneeth Vepakomma, Ramesh Raskar, Mehdi Bennis - Topic: Split Learning, Differential Privacy (2022) @INTERPOLATE — Workshop on Interpolation Regularizers and Beyond-NeurIPS
30. Formal Privacy Guarantees for Neural Network queries by estimating local Lipschitz constant (PDF), @Workshop on Formal Verification of Machine Learning-ICML, Abhishek Singh, Praneeth Vepakomma, Vivek Sharma, Ramesh Raskar - Topic: Differential Privacy, Modified Propose-Test-Release, Lipschitz Constant (2022)
29. Decouple-and-Sample: Protecting sensitive information in task agnostic data release, (PDF) @ECCV 2022, European Conference on Computer Vision, (ECCV 2022) Abhishek Singh, Ethan Garza, Ayush Chopra, Praneeth Vepakomma, Vivek Sharma, Ramesh Raskar, -Topic: Synthetic data, Differential privacy (only on partial component of latent space-a double-edged sword), Decorrelation (2022)
28. Private measurement of nonlinear correlations between data hosted across multiple parties, Praneeth Vepakomma, Subha Nawer Pushpita, Ramesh Raskar (PDF) -Topic: Statistics, Differential Privacy (2021)
27. PrivateMail: Differentially private supervised manifold learning of deep features with privacy, @AAAI 2022, 36th AAAI Conference on Artificial Intelligence, (AAAI 2022) (Oral), Praneeth Vepakomma, Julia Balla, Ramesh Raskar (PDF) -Topic: Differential Privacy, Privacy Preserving ML, On-device ML (2022)
26. LocFedMix-SL: Localize, Federate, and Mix for Improved Scalability, Convergence, and Latency in Split Learning, @ WWW 2022 : International World Wide Web Conference/The Web Conf (WWW 2022), Seungeun Oh, Jihong Park, Praneeth Vepakomma, Sihun Baek, Ramesh Raskar, Mehdi Bennis and Seong-Lyun Kim (PDF) Distributed ML, Split Learning
25. An Automated Framework for Distributed Deep Learning--A Tool Demo (Link for the Demo), Gharib Gharibi, Ravi Patel, Anissa Khan, Babak Poorebrahim Gilkalaye, Praneeth Vepakomma, Ramesh Raskar, Steve Penrod, Greg Storm, Riddhiman Das∗ Topic: Demo, @ICDCS 2022, 42nd IEEE International Conference on Distributed Computing Systems, (ICDCS 2022) (2022)
24. DISCO: Dynamic and invariant sensitive channel obfuscation for deep neural networks, @CVPR 2021, IEEE Computer Vision and Pattern Recognition Conference, A.Singh, A.Chopra,V.Sharma, E.Garza, E.Zhang, P.Vepakomma, R.Raskar-Main Paper: (PDF) on -Topic: Preventing reconstruction attacks, Distributed Inference, (CVPR 2021)
23. (New!) NoPeek-Infer: Preventing face reconstruction attacks in distributed inference after on-premise training, @FG 2021, IEEE International Conference on Automatic Face and Gesture Recognition (IEEE FG 2021) (Oral) , Praneeth Vepakomma, Abhishek Singh, Emily Zhang, Otkrist Gupta, Ramesh Raskar, (PDF) -Topic: Reconstruction attacks, ML (2021) (Mukh Best Paper Runner Up Award at IEEE FG 2021)
22. Supervised Dimensionality Reduction via Maximization of Distance Correlation, @Electronic Journal of Statistics, (Journal), P.Vepakomma, C. Tonde and A.Elgammal, (PDF) -Topic: Statistics, Optimization, ML
21. FedML: A research library and benchmark for federated machine learning C He, S Li, J So, M Zhang, H Wang, X Wang, P Vepakomma, A Singh, R Raskar (PDF) (Best Paper Award at NeurIPS 2020-SpicyFL), -Topic: Federated ML/Distributed ML/Systems (2021)
20. AirMixML: Over-the-Air Data Mixup for Inherently Privacy-Preserving Edge Machine Learning, @(GLOBECOM 2021), IEEE Global Communications Conference, Yusuke Koda, Jihong Park, Mehdi Bennis, Praneeth Vepakomma, Ramesh Raskar (PDF), – Topic: Differential Privacy, ML, Wireless, Analog communication, Over-the-air, Early proof of concept (Early PoC)
19. AdaSplit: Adaptive trade-offs for resource-constrained distributed deep learning, Ayush Chopra, Surya Kant Sahu, Abhishek Singh, Abhinav Java, Praneeth Vepakomma, Vivek Sharma, Ramesh Raskar (PDF) -Topic: Distributed ML/Split Learning (2022)
18. Diverse data selection via combinatorial quasi-concavity of distance covariance: A polynomial time global minimax algorithm, (PDF), Praneeth Vepakomma, Yulia Kempner, @Discrete Applied Mathematics (Journal). -Topic: Statistics, optimization (2019)
17. Advances and open problems in federated learning, (PDF) @Foundations and Trends in Machine Learning, Vol 14, Issue 1–2 with 58 authors from 25 institutions, (Journal), -Topic: Federated ML/Distributed ML (2021)
16. Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity, Praneeth Vepakomma, Yulia Kempner, Ramesh Raskar, @SubSetML: Subset Selection in Machine Learning: From Theory to Practice- ICML 2021 Workshop (PDF) -Topic: Distributed ML, Combinatorial statistics, Combinatorial optimization (2021)
15. D.A.M.S: Meta-estimators of private sketch data structures for differentially private COVID-19 contact tracing, (PDF), P.Vepakomma, S.N. Pushpita, R. Raskar, PRIML AND PPML JOINT EDITION @ NeurIPS-2020 Workshop, -Topic: Statistics, Differential Privacy, Privacy Preserving ML (2020)
14. Split learning for health: Distributed deep learning without sharing raw patient data, (PDF) Praneeth Vepakomma, Otkrist
Gupta, Tristan Swedish, Ramesh Raskar https://arxiv.org/pdf/1812.00564.pdf, Project page: https://splitlearning.github.io/, @ICLR 2019 Workshop on AI for social good. -Topic: Federated ML/Distributed ML, Split Learning (2018)
13. A Fast Algorithm for Manifold Learning by Posing it as a Symmetric Diagonally Dominant Linear System, (PDF), Praneeth
Vepakomma & Ahmed Elgammal, Applied and Computational Harmonic Analysis, (Journal) – Topic: ML
12. Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning, @ FL-AAAI 2022 Workshop, Shraman Pal, Mansi Uniyal, Jihong Park, Praneeth Vepakomma, Ramesh Raskar, Mehdi Bennis, Moongu Jeon, Jinho Choi (PDF) -Topic: Distributed ML, Split Learning (2022)
11. Splintering with distributions: A stochastic decoy scheme for private computation, P Vepakomma, J Balla, R Raskar, arXiv preprint arXiv:2007.02719 -Topic: Privacy Preserving ML
10. SplitNN-driven Vertical Partitioning, (PDF) I Ceballos, V Sharma, E Mugica, A Singh, A Roman, P Vepakomma, arXiv preprint arXiv:2008.0413, -Topic: Federated ML/Distributed ML
9. Privacy in Deep Learning: A Survey, (PDF) F Mirshghallah, M Taram, P Vepakomma, A Singh, R Raskar-Topic: Privacy Preserving ML
8. PPContactTracing: A Privacy-Preserving Contact Tracing Protocol for COVID-19 Pandemic, (PDF) P Singh, A Singh, G
Cojocaru, P Vepakomma, R Raskar arXiv preprint arXiv:2008.06648-Topic: Cryptographic Tools
7. Assessing Disease Exposure Risk with Location Histories and Protecting Privacy: A Cryptographic Approach in
Response to A Global Pandemic, (PDF) A Berke, M Bakker, P Vepakomma, R Raskar, K Larson, AS Pentland-Topic: Cryptographic Tools
6. ExpertMatcher: Automating ML Model Selection for Clients using Hidden Representations, (PDF) Vivek Sharma, Praneeth
Vepakomma, Tristan Swedish, Ken Chang, Jayashree Kalpathy-Cramer, and Ramesh Raskar, NeurIPS Workshop on
Robust AI in Financial Services: Data, Fairness, Explainability, Trustworthiness, and Privacy, -Topic: Federated ML/Distributed ML
5. A Review of Homomorphic Encryption Libraries for Secure Computation, (PDF) Sai Sri Sathya, Praneeth Vepakomma,
Ramesh Raskar, Ranjan Ramachandra, and Santanu Bhattacharya, -Topic: Cryptographic Tools
4. Optimal bandwidth estimation for a fast manifold learning algorithm to detect circular structure in highdimensional
data, (PDF) Praneeth Vepakomma and Susovan Pal, Topic: ML
3. Scoring practices for remote sensing of land mines (PDF), Praneeth Vepakomma, Adel Al Weshah, Aaron Bardall, Valeria
Barra, Dean Duffy, Hamza Ghadgali, Sarafa Iyaniwura , Hangjie Ji, Qingxia Li, Richard Moore, Kenneth Morton, Ryan
Pellico, Christina Selby, Razvan Stefanescu, Melissa Strait, Zhe Wang, Andres Vargas, Mathematical Problems in
Industry, Duke University-Department of Mathematics, -Topic: ML
2. A-Wristocracy: Deep Learning on Wrist-worn Sensing for Recognition of User Complex Activities, (PDF), Praneeth
Vepakomma, Debraj De, Sajal K. Das & Shekhar Bhansali at IEEE Body Sensor Networks Conference, at MIT Media Lab,
Cambridge, IEEE-BSN 2015, -Topic: ML, Wearables, Hardware/Software Integration
1. Split Learning on FPGA’s and FPIA’s: The first Field Programmable Imaging Arrays, Hannah Whisnant, Praneeth
Vepakomma, Ramesh Raskar, -Topic: Federated ML, On-Device ML, Hardware/Software Integration
Organizer:
- Organizing: NeurIPS International Workshop on Federated Learning: Recent Advances and New Challenges in Conjunction (FL-NeurIPS'22)
- Organized: ICML International Workshop on Federated Learning for User Privacy and Data Confidentiality, 2021
- Keynote Speakers: Sebastian U. Stich, Nic Lane, Ramesh Raskar, Ameet Talwalkar, Filip Hanzely, Dimitris Papailiopoulos and Salman Avestimehr
- Organized: ICLR Workshop on Distributed and Private Machine Learning (DPML, 2021).
- Keynote Speakers: David Evans, Lalitha Sankar, Gauri Joshi & Graham Cormode.
- CVPR Tutorial On Distributed Private Machine Learning for Computer Vision: Federated Learning, Split Learning and Beyond, 2019
- Tutorial Speakers: Brendan McMahan, Jakub Konečný, Ramesh Raskar, Praneeth Vepakomma, Otkrist Gupta, Hassan Takabi
- Organized: Workshop on Split Learning for Distributed Machine Learning (SLDML’21).
- Keynote Speakers: Peter Kairouz, OpenMined, Geeta Chauhan and Supriyo Chakraborty.
Community Service:
- Volunteered and conducted interactive demos for students at Shiloh Point Elementary School, Cumming, GA (Atlanta) for the Family Science Night.
- Service at homeless shelter in Salt Lake City, UT with PublicEngines. Helped with sorting of donation material (non-monetary).
Bio: Praneeth Vepakomma is currently a PhD student at MIT in the Camera Culture group and works on distributed and private computation. He has extensive industrial experience across Meta, Apple, Amazon, Motorola Solutions, Corning and several startups. His research focuses on developing algorithms for distributed computation in statistics & machine learning under constraints of privacy, & efficiency. He won the Meta PhD research fellowship in Applied Statistics, two SERC Scholarships (for Social and Ethical Responsibilities of Computing) from MIT's Schwarzman college of computing. He co-founded a research based non-profit (Integrity Distributed) that won the Financial Times Digital Innovation Award. He won a Best Student Paper Award at FL-IJCAI, a Baidu Best Paper Award at NeurIPS-SpicyFL and a Best Paper Runner Up Award at FG-2021. His technical work is inspired by foundations of non-asymptotic statistics, randomized algorithms, learning augmented algorithms, combinatorics, and at times just by systems design. He was Interviewed in the book, 'Data Scientist: The Definitive Guide to Becoming a Data Scientist' and has organized several workshops at ICLR, ICML, IJCAI, CVPR and NeurIPS.