Scholar
Virat Shejwalkar
Google Scholar ID: M6GAEdUAAAAJ
Google DeepMind
Citations & Impact
Citations
2,382
H-index
13
i10-index
15
Publications
20
Co-authors
17
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - 'The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning', IEEE/CVF International Conference on Computer Vision (ICCV), 2023
- - 'Recycling Scraps: Improving Private Learning Using Intermediate Checkpoints', AAAI Privacy Preserving Artificial Intelligence (PPAI) Workshop, 2023
- - 'On The Pitfalls of Security Evaluation of Robust Federated Learning', Deep Learning Security and Privacy Workshop at IEEE S&P, 2023
- - 'Every Vote Counts: Ranking-Based Training of Federated Learning to Resist Poisoning Attacks', USENIX Security Symposium, 2023
- - More publications available in the original text.
Research Experience
- Before joining the PhD program, worked at the CryptoLux group of the University of Luxembourg, developing FELICS, a performance benchmarking tool for lightweight cryptography hardware, under the guidance of Professor Alex Biryukov.
Education
- PhD: University of Massachusetts Amherst, advised by Professor Amir Houmansadr. Undergraduate: IIT Bombay, thesis on countermeasures against side channel attacks on AES hardware, under the guidance of Professor Virendra Singh.
Background
- Research interests: Security and privacy of machine learning. Specific areas include: 1) Exploiting and fixing the vulnerability of federated learning to various types of poisoning threats; 2) Developing private information inference attacks and defenses for centralized and distributed learning algorithms.
