I am an assistant professor in the Department of Electrical and Systems Engineering (as of July 2017). I hold secondary appointments in the Department of Computer and Information Systems as well as the Department of Statistics and Data Science at the Wharton Business School. I am also a faculty affiliate of the Warren Center for Network and Data Sciences.
Before joining Penn, I was a research fellow at the Simons Institute, UC Berkeley (program: Foundations of Machine Learning). Prior to that, I was a post-doctoral scholar and lecturer in the Institute for Machine Learning at ETH Zürich. I received my Ph.D. degree in Computer and Communication Sciences from EPFL.
Some Recent Publications
• A.Robey, L. Chamon, G. Pappas, H. Hassani Probabilistically Robust Learning: Balancing Average- and Worst-case Performance, 2022.
• H. Hassani, A. Javanmard The curse of overparametrization in adversarial training: Precise analysis of robust generalization for random features regression, 2022.
• A. Zhou, F. Tajwar, A. Robey, T. Knowles, G. Pappas, H. Hassani, C. Finn Do Deep Networks Transfer Invariances Across Classes?, 2022.
• A. Robey, G. Pappas, H. Hassani Model-Based Domain Generalization, 2021.
• A. Adibi, A. Mokhtari, H. Hassani Minimax Optimization: The Case of Convex-Submodular
, 2021.
• L. Collins, H. Hassani, A. Mokhtari, S. Shakkottai Exploiting Shared Representations for Personalized Federated Learning, 2021.
• E. Lei, H. Hassani, S. Saeedi Bidokhti Out-of-Distribution Robustness in Deep Learning Compression, 2021.
• P. Delgosha, H. Hassani, R. Pedarsani Robust Classification Under L_0 Attack for the Gaussian Mixture Model, 2021.
• A. Mitra, R. Jaafar, G. Pappas, H. Hassani Achieving Linear Convergence in Federated Learning under Objective and Systems Heterogeneity, 2021.
• Z. Shen, H. Hassani, S. Kale, A. Karbasi Federated Functional Gradient Boosting, 2021.
• Z. Shen, Z. Wang, A. Ribeiro, H. Hassani Sinkhorn Natural Gradient for Generative Models, 2020.
• A. Reisizadeh, I. Tziotis , H. Hassani, A. Mokhtari, R. Pedarsani Straggler-Resilient Federated Learning:
Leveraging the Interplay Between Statistical Accuracy
and System Heterogeneity, 2020.
• A. Robey, H. Hassani, G. J. Pappas Model-Based Robust Deep Learning, 2020.
• A. Javanmard, M. Soltanolkotabi, H. Hassani Precise Tradeoffs in Adversarial Training for Linear Regression, 2020.
• Z. Shen, Z. Wang, A. Ribeiro, H. Hassani Sinkhorn Barycenter via Functional Gradient Descent, 2020.
• A. Adibi, A. Mokhtari, H. Hassani Submodular Meta-Learning, 2020.
• E. Dobriban, H. Hassani, D. Hong, A. Robey Provable tradeoffs in adversarially robust classification
, 2020.
• X. Chen, K. Gatsis, H. Hassani and S. Saeedi Age of Information in Random Access Channels, 2020.
• H. Hassani, A. Karbasi, A. Mokhtari, Z. Shen Stochastic Conditional Gradient++, 2019.
• M. Fazlyab, A. Robey, H. Hassani, M. Morari, G. Pappas Efficient and Accurate Estimation of Lipschitz Constants for Deep Neural Networks, 2019.
• A. Robey, A. Adibi, B. Schlotfeldt, G. Pappas, H. Hassani Optimal Algorithms for Submodular Maximization with Distributed Constraints, 2019.
• A. Reisizadeh, H. Taheri, A. Mokhtari, H. Hassani, R. Pedarsani Robust and Communication-Efficient Collaborative Learning, 2019.
• Z. Shen, H. Hassani, A. Ribeiro Hessian Aided Policy Gradient, 2019.
• M. Zhang, L. Chen, A. Mokhtari, H. Hassani, A. Karbasi Quantized Frank-Wolfe: Communication-Efficient Distributed Optimization, 2019.
•A. Gotovos, H. Hassani, A. Krause, S. Jegelka, Discrete Sampling Using Semigradient-based Product Mixtures, 2018.
• A. Mokhtari, H. Hassani, A. Karbasi, Stochastic Conditional Gradient Methods: From Convex Minimization to Submodular Maximization, 2018.
• Y. Balaji, H. Hassani, R. Chellappa, S. Feizi, Entropic GANs meet VAEs: A Statistical Approach to Compute Sample Likelihoods in GANs, 2018.
• K. Gatsis, H. Hassani, G. J. Pappas, Latency-Reliability Tradeoffs for State Estimation, 2018.
• A. Mokhtari, H. Hassani, A. Karbasi, Decentralized Submodular Maximization: Bridging Discrete and Continuous Settings, 2018.
• M. Fereydounian, V. Jamali, H. Hassani, H. Mahdavifar, Channel Coding at Low Capacity, 2018.
• A. Fazeli, H. Hassani, M. Mondelli, A. Vardy, Binary Linear Codes with Optimal Scaling: Polar Codes with Large Kernels, 2018.
• H. Hassani, S. Kudekar, O. Ordentlich, Y. Polyanskiy, R. Urbanke, Almost Optimal Scaling of Reed-Muller Codes on BEC and BSC Channels, 2018.
• L. Chen, C. Harshaw, H. Hassani, A. Karbasi, Projection-Free Online Optimization with Stochastic Gradient: From Convexity to Submodularity, 2018.
• M. Hayhoe, F. Barreras, H. Hassani, V. M. Preciado, SPECTRE: Seedless Network Alignment via Spectral Centralities, 2018.
• Y. Chen, S. H. Hassani, A. Krause, Near-optimal Bayesian Active Learning with Correlated and Noisy Tests, 2017.
News
• Honored to be selected as a Distinguished Lecturer of the IEEE Information Theory Society in 2022-2023.
• Invited talks at UT Austin (Foundations of Data Science Seminar), Chalmers University (ML seminar), MIT (OPT-ML++ Seminar), Rutgers University (Business School MSIS Seminar), University of Wisconsin-Maddison (SILO Seminar), University of Washington Seattle (ML Seminar) Video. 2021.
• NEURIPS 2021: We will present our works on Model Based Domain Generalization and Linear Convergence in Federated Learning Under Statistical and Computational Heterogeneity and Adversarial robustness with semi-infinite constrained learning
• ICML 2021: We will present our work on Exploiting Shared Representations for Federated Learning under Heterogeneity
• Honored to be selected among "Intel’s 2020 Rising Star Faculty Awardees".
• NEURIPS 2020: We will present our works on Submodular Meta-Learning and Efficiently Computing Sinkhorn Barycenters and A Natural Gradient Method to Compute The Sinkhorn Distance in Optimal Transport (as a Spotlight)
• COLT 2020: We will present our work on Precise Tradesoffs for Adversarial Training
• ICML 2020: We will present our work on Decenteralized Optimization over Directed Networks
• NSF CAREER Award, 2019.
• AISTATS 2020: We will present our works on Federated Learning and Quantized and Distributed Frank-Wolfe and Black-Box Submodular Maximization and One Sample Frank-Wolfe.
• AFOSR Young Investigator Award, 2019.
• NEURIPS 2019: We will present our works on Stochastic Conditional Gradient++ and Efficient and Accurate Estimation of Lipschitz Constants for Deep Neural Networks and Robust and Communication-Efficient Collaborative Learning and Bandit Submodular Maximization.
• ICML 2019: We will present our works on Hessian-Aided Policy Gradient and A Statistical Approach to Compute Sample Likelihoods in GANs.
• Data Science Summer School, Ecole Polytechnique, 2019: I will be giving 6 lectures on "Theory and Applications of Submodularity: From Discrete to Continuous and Back".
• Allerton 2018: We will be hoding a session on "Submodular Optimization" with 5 great speakers.
• ICML 2018: We will present our work on Decenteralized Submodular Maximization and Projection-Free Online Optimization.
• ISIT 2018: We will present our work on The Scaling of Reed-Muller Codes and A New Coding Paradigm for the Primitive Relay Channel.
• Invited talk at the workshop on local algorithms (MIT) on "Submodular Maximization: The Decentralized Setting" (June 15th).
• AISTATS 2018: We will present our work on The Stochastic Frank-Wolfe Method and Online Submodular Maximization.
• Invited talk at the workshop on coding and information theory (Harvard, CSMA) and the University of Maryland (ECE) on "Non-asymptotic Analysis of Codes and its Practical Significance" (April 13th and March 29th).
• NSF CISE Research Initiative (NSF-CRII) award, 2018.
• Invited talk at the Santa Fe Institute on "Sequential Information Maximization: From Theory to Designs" (Feb 21st).
• Talk at the Dagstuhl Seminar and ITA 2018 on "Decentralized Submodular Maximization: Bridging Discrete and Continuous Settings" (Feb 16th).
• Talk at UPenn ESE on "Coding for IoT" (Jan 26th).
• We are organizing a session on "Submodular Optimization" at the 2018 INFORMS Optimization Society Conference.
• AAAI 2018: We will present our work Learning to Interact with Learning Agents.
• I will serve as a program committe member for IEEE International Symposium on Information Theory (ISIT'18). Please consider submitting your work to ISIT!
• NIPS 2017: We will present our works on Stochastic Submodular Maximization and Gradient Methods for Submodular Maximization.
• Invited talk at MIT EECS on "Recent Advances in Channel Coding" (Nov 1st).
• Invited talk at Yale Institute for Networking Science (YINS) on "K-means: A Nonconvex problem with Fast and Provable Algorithms" (Oct 25th).
Contact
In person : | 465C (3401 Walnut st.) | |
Cell : | 650 666 5254 | |
email : | hassani@seas.upenn.edu | |
mail: | Dept. of Electrical & Systems Engineering | |
University of Pennsylvania | ||
Room 465C | ||
3401 Walnut Street | ||
Philadelphia, PA 19104 |