7 papers
Joint Domain-Class Modeling for Federated Learning Under Feature Skew
Sina Najafi, Mostafa Tavassolipour, Seyed Pooya Shariatpanahi
Federated learning (FL) enables collaborative model training without centralizing private data, but performance often degrades under feature skew: clients share labels while the co…
Domain-Aware Pruning: Sparsity and Domain Generalization via Regularized Probabilistic Masking
Parham Sazdar, Mostafa Tavassolipour, Reshad Hosseini
Domain generalization (DG) and neural network pruning are conventionally treated as distinct objectives, targeting out-of-distribution (OOD) robustness and model efficiency, respec…
Improving Adversarial Robustness of Zero-Shot CLIP with Confidence-Aware Weighting
Nikoo Naghavian, Mostafa Tavassolipour
Vision-language models such as CLIP demonstrate impressive zero-shot generalization but remain highly vulnerable to adversarial attacks. Prior adversarial methods treat all samples…
Robustifying Diffusion-Denoised Smoothing Against Covariate Shift
Ali Hedayatnia, Mostafa Tavassolipour, Babak Nadjar Araabi +1
Randomized smoothing is a well-established method for achieving certified robustness against l2-adversarial perturbations. By incorporating a denoiser before the base classifier, p…
FedRP: A Communication-Efficient Approach for Differentially Private Federated Learning Using Random Projection
Sina Najafi, Mostafa Tavassolipour, Mohammad Hasan Narimani
Federated learning (FL) enables collaborative model training without centralizing data, but exchanging high-dimensional updates can expose sensitive information and incur substanti…
Out-of-distribution detection using normalizing flows on the data manifold
Seyedeh Fatemeh Razavi, Mohammad Mahdi Mehmanchi, Reshad Hosseini +1
Using the intuition that out-of-distribution data have lower likelihoods, a common approach for out-of-distribution detection involves estimating the underlying data distribution.…
Joint Manifold Learning and Density Estimation Using Normalizing Flows
Seyedeh Fatemeh Razavi, Mohammad Mahdi Mehmanchi, Reshad Hosseini +1
Based on the manifold hypothesis, real-world data often lie on a low-dimensional manifold, while normalizing flows as a likelihood-based generative model are incapable of finding t…