Virat Shejwalkar

30 papers A* 6A 2C 2Misc 2Journal 17Unranked 1

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Decoding FL Defenses: Systemization, Pitfalls, and Remedies. CoRR Momin Ahmad Khan, Virat Shejwalkar, Yasra Chandio, Amir Houmansadr, Fatima Muhammad Anwar
2025	C	`conf`	Improving Streaming ASR via Differentially Private Fusion of Data from Multiple Sources. ASRU Virat Shejwalkar, Om Thakkar, Steve Chien, Nicole Rafidi, Arun Narayanan
2025	J	`jnl`	Multilingual and Multi-Accent Jailbreaking of Audio LLMs. CoRR Jaechul Roh, Virat Shejwalkar, Amir Houmansadr
2025	J	`jnl`	Rebellion: Noise-Robust Reasoning Training for Audio Reasoning Models. CoRR Tiansheng Huang, Virat Shejwalkar, Oscar Chang, Milad Nasr, Ling Liu
2025	J	`jnl`	Recycling Scraps: Improving Private Learning by Leveraging Checkpoints]{Recycling Scraps: Improving Private Learning by Leveraging Checkpoints. Proc. Priv. Enhancing Technol. Virat Shejwalkar, Arun Ganesh, Rajiv Mathews, Yarong Mu, Shuang Song, Om Thakkar, Abhradeep Thakurta, Xinyi Zheng
2024	A	`conf`	Quantifying Unintended Memorization in BEST-RQ ASR Encoders. INTERSPEECH Virat Shejwalkar, Om Thakkar, Arun Narayanan
2023	A*	`conf`	Every Vote Counts: Ranking-Based Training of Federated Learning to Resist Poisoning Attacks. USENIX Security Symposium Hamid Mozaffari, Virat Shejwalkar, Amir Houmansadr
2023	—	`conf`	On the Pitfalls of Security Evaluation of Robust Federated Learning. SP (Workshops) Momin Ahmad Khan, Virat Shejwalkar, Amir Houmansadr, Fatima M. Anwar
2023	A*	`conf`	The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning. ICCV Virat Shejwalkar, Lingjuan Lyu, Amir Houmansadr
2022	A*	`conf`	Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Production Federated Learning. SP Virat Shejwalkar, Amir Houmansadr, Peter Kairouz, Daniel Ramage
2022	J	`jnl`	Machine Learning with Differentially Private Labels: Mechanisms and Frameworks. Proc. Priv. Enhancing Technol. Xinyu Tang, Milad Nasr, Saeed Mahloujifar, Virat Shejwalkar, Liwei Song, Amir Houmansadr, Prateek Mittal
2022	A*	`conf`	Mitigating Membership Inference Attacks by Self-Distillation Through a Novel Ensemble Architecture. USENIX Security Symposium Xinyu Tang, Saeed Mahloujifar, Liwei Song, Virat Shejwalkar, Milad Nasr, Amir Houmansadr, Prateek Mittal
2022	J	`jnl`	Recycling Scraps: Improving Private Learning by Leveraging Intermediate Checkpoints. CoRR Virat Shejwalkar, Arun Ganesh, Rajiv Mathews, Om Thakkar, Abhradeep Thakurta
2022	Misc	`conf`	Security Analysis of SplitFed Learning. SenSys Momin Ahmad Khan, Virat Shejwalkar, Amir Houmansadr, Fatima M. Anwar
2022	J	`jnl`	Security Analysis of SplitFed Learning. CoRR Momin Ahmad Khan, Virat Shejwalkar, Amir Houmansadr, Fatima M. Anwar
2022	J	`jnl`	The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning. CoRR Virat Shejwalkar, Lingjuan Lyu, Amir Houmansadr
2022	Misc	`conf`	Towards privacy aware deep learning for embedded systems. SAC Vasisht Duddu, Antoine Boutet, Virat Shejwalkar
2021	J	`jnl`	Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Federated Learning. CoRR Virat Shejwalkar, Amir Houmansadr, Peter Kairouz, Daniel Ramage
2021	J	`jnl`	FSL: Federated Supermask Learning. CoRR Hamid Mozaffari, Virat Shejwalkar, Amir Houmansadr
2021	A*	`conf`	Manipulating the Byzantine: Optimizing Model Poisoning Attacks and Defenses for Federated Learning. NDSS Virat Shejwalkar, Amir Houmansadr
2021	A*	`conf`	Membership Privacy for Machine Learning Models Through Knowledge Transfer. AAAI Virat Shejwalkar, Amir Houmansadr
2021	J	`jnl`	Mitigating Membership Inference Attacks by Self-Distillation Through a Novel Ensemble Architecture. CoRR Xinyu Tang, Saeed Mahloujifar, Liwei Song, Virat Shejwalkar, Milad Nasr, Amir Houmansadr, Prateek Mittal
2020	J	`jnl`	GECKO: Reconciling Privacy, Accuracy and Efficiency in Embedded Deep Learning. CoRR Vasisht Duddu, Antoine Boutet, Virat Shejwalkar
2020	J	`jnl`	Leveraging Prior Knowledge Asymmetries in the Design of Location Privacy-Preserving Mechanisms. IEEE Wirel. Commun. Lett. Nazanin Takbiri, Virat Shejwalkar, Amir Houmansadr, Dennis L. Goeckel, Hossein Pishro-Nik
2020	C	`conf`	Quantifying Privacy Leakage in Graph Embedding. MobiQuitous Vasisht Duddu, Antoine Boutet, Virat Shejwalkar
2020	J	`jnl`	Quantifying Privacy Leakage in Graph Embedding. CoRR Vasisht Duddu, Antoine Boutet, Virat Shejwalkar
2019	J	`jnl`	Cronus: Robust and Heterogeneous Collaborative Learning with Black-Box Knowledge Transfer. CoRR Hongyan Chang, Virat Shejwalkar, Reza Shokri, Amir Houmansadr
2019	J	`jnl`	Leveraging Prior Knowledge Asymmetries in the Design of Location Privacy-Preserving Mechanisms. CoRR Nazanin Takbiri, Virat Shejwalkar, Amir Houmansadr, Dennis L. Goeckel, Hossein Pishro-Nik
2019	J	`jnl`	Reconciling Utility and Membership Privacy via Knowledge Distillation. CoRR Virat Shejwalkar, Amir Houmansadr
2019	A	`conf`	Revisiting utility metrics for location privacy-preserving mechanisms. ACSAC Virat Shejwalkar, Amir Houmansadr, Hossein Pishro-Nik, Dennis Goeckel