no code implementations • 13 Feb 2024 • Mohammad Mehrabi, Stefan Wager
In this paper, we re-visit the task of off-policy evaluation in MDPs under a weaker notion of distributional overlap, and introduce a class of truncated doubly robust (TDR) estimators which we find to perform well in this setting.
no code implementations • 20 Jun 2023 • Mohammad Mehrabi, Ryan A. Rossi
Ideally, it is desired to understand how a set of collected features combine together and influence the response value, but this problem is notoriously difficult, due to the high-dimensionality of data and limited number of labeled data points, among many others.
no code implementations • 5 Sep 2022 • Adel Javanmard, Mohammad Mehrabi
Performance of classifiers is often measured in terms of average accuracy on test data.
1 code implementation • 22 Oct 2021 • Adel Javanmard, Mohammad Mehrabi
We develop a theory to show that the low-dimensional manifold structure allows one to obtain models that are nearly optimal with respect to both, the standard accuracy and the robust accuracy measures.
no code implementations • 5 Mar 2021 • Mohammad Mehrabi, Aslan Tchamkerten
Experimental results with random Gaussian design matrices show that LiRE substantially reduces the number of measurements needed for perfect support recovery via Compressive Sampling Matching Pursuit, Basis Pursuit (BP), and OMP.
Information Theory Information Theory
no code implementations • 15 Jan 2021 • Mohammad Mehrabi, Adel Javanmard, Ryan A. Rossi, Anup Rao, Tung Mai
We study the tradeoff between standard risk and adversarial risk and derive the Pareto-optimal tradeoff, achievable over specific classes of models, in the infinite data limit with features dimension kept fixed.
no code implementations • 4 Nov 2019 • Yash Deshpande, Adel Javanmard, Mohammad Mehrabi
Adaptive collection of data is commonplace in applications throughout science and engineering.
no code implementations • ICML 2018 • Mohammad Mehrabi, Aslan Tchamkerten, Mansoor I. Yousefi
The approximation power of general feedforward neural networks with piecewise linear activation functions is investigated.