Kareem Amin

Bio

I joined Google Research in 2016, where I have spent most of my time working on problems in the intersection of privacy, machine learning and AI, including the creation of privacy-enhancing technology and infrastructure. More broadly, my research interests have tended to focus on learning when there is strategic or adversarial reasoning involved. Prior to joining Google I was a research fellow at the University of Michigan, hosted by professors Satinder Singh and Michael Wellman, where I worked on security games. I received my PhD at the University of Pennsylvania, where I was advised by professor Michael Kearns, and published on the foundations of machine learning with a focus on online learning and applications to auctions, advertising and markets.

Papers

• Amin, K., Bie, A., Kong, W., Syed, U., Vassilvitskii, S. Learning from Synthetic Data: Limitations of ERM. The 37th International Conference on Algorithmic Learning Theory (ALT 2026).
• Amin, K., Babakniya, S., Bie, A., Kong, W., Syed, U., Vassilvitskii, S. Escaping Collapse: The Strength of Weak Data for Large Language Model Training. 39th Conference on Neural Information Processing Systems (NeurIPS 2025).
• Amin, K., Kulesza, K., Vassilvitskii S. Practical Considerations for Differential Privacy. Theory and Practice of Differential Privacy (TPDP 2024).
• Amin, K., Bie, A., Kong, W., Kurakin, A., Ponomareva, N., Syed, U., Terzis, A., Vassilvitskii S. Private Prediction for Large-Scale Synthetic Text Generation. Findings of the Association for Computational Linguistics. (EMNLP 2024) .
• Khodak, M., Amin, K., Dick, T., Vassilvitskii, S. Learning-augmented Private Algorithms for Multiple Quantile Release. Proceedings of the 40th International Conference on Machine Learning (ICML 2023).
• Amin, K., Joseph, M., Ribero, M., Vassilvitskii, S. Easy Differentially Private Linear Regression. The Eleventh International Conference on Learning Representations (ICLR 2023).
• Amin, K., DeSalvo, G., Rostamizadeh, A. Learning with Labeling Induced Abstentions. 35th Conference on Neural Information Processing Systems (NeurIPS 2021).
• Levy, D., Sun, Z., Amin, K., Kale, S., Kulesza, A., Mohri, M., Suresh., A. Learning with User-Level Privacy. 35th Conference on Neural Information Processing Systems (NeurIPS 2021).
• Amin, K., Joseph, M., Mao, J. Pan-Private Uniformity Testing. Thirty-Third Annual Conference on Learning Theory (COLT 2020).
• Amin, K., Cortes, C., DeSalvo, G., Rostamizadeh, A. Understanding the Effects of Batching in Online Active Learning. The 23rd International Conference on Artificial Intelligence and Statistics (AISTATS 2020).
• Amin, K., Dick, T., Kulesza, A., Medina, A., Vassilvitskii, S. Differentially Private Covariance Estimation. 33rd Conference on Neural Information Processing Systems (NeurIPS 2019).
• Amin, K., Kulesza, A., Medina, A., Vassilvitskii, S. Bounding User Contributions: A Bias-Variance Trade-off in Differential Privacy. Proceedings of the 31st International Conference on Machine Learning (ICML 2019).
• Amin, K., Jiang, N., Singh, S. Repeated Inverse Reinforcement Learning. Advances in Neural Information Processing Systems 30 (NIPS 2017).
• Abernethy, J., Amin, K., Zhu, R. Threshold Bandit, With and Without Censored Feedback. Advances in Neural Information Processing Systems 29 (NIPS 2016) .
• Venkatesan, S., Albanese M., Amin, K., Jajodia, S., Wright, M. A Moving Target Defense Approach to Mitigate DDoS Attacks Against Proxy-Based Architectures. IEEE Conference on Communications and Network Security (CNS 2016) .
• Cheng, F., Liu, J., Amin, K., Wellman, M. Strategic Payment Routing in Financial Credit Networks. Seventeenth ACM Conference on Economics and Computation (EC 2016) .
• Amin, K., Singh S., Wellman M. Gradient Methods for Stackelberg Security Games. Uncertainty in Aritifical Intelligence : Proceedings of the Thirty-Second Conference (UAI 2016). Stackgrad Implementation (in Python).
• Amin, K., Cummings, R., Dworkin, L., Kearns, M., Roth, A. Online Learning and Profit Maximization from Revealed Preferences. Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI 2015).
• Amin, K., Kale, S., Tesauro, G., Turaga, D. Budgeted Prediction With Expert Advice. Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI 2015).
• Amin, K., Rostamizadeh, A. Syed, U. Repeated Contextual Auctions with Strategic Buyers. Advances in Neural Information Processing Systems 27 (NIPS 2014).
• Amin, K., Heidari H., Kearns M. Learning from Contagion (Without Timestamps). Proceedings of the 31st International Conference on Machine Learning (ICML 2014).
• Amin, K., Rostamizadeh, A., Syed, U. Learning Prices for Repeated Auctions with Strategic Buyers. Advances in Neural Information Processing Systems 26 (NIPS 2013).
• Abernethy, J., Amin, K., Draief, M., Kearns, M. Large-Scale Bandit Problems and KWIK Learning. Proceedings of the 30th International Conference on Machine Learning (ICML 2013).
• Amin, K., Kearns, M., Key, P., Schwaighofer, A. Budget Optimization for Sponsored Search: Censored Learning in MDPs. Uncertainty in Artificial Intelligence: Proceedings of the Twenty-Eighth Conference (UAI 2012).
• Amin, K., Kearns, M., Syed, U. Graphical models for bandit problems. Uncertainty in Artificial Intelligence: Proceedings of the Twenty-Seventh Conference (UAI 2011).
• Amin, K., Kearns, M., Syed, U. Bandits, query learning, and the haystack dimension. Proceedings of the Twenty-Fourth Annual Conference on Learning Theory (COLT 2011).