Publications de Rahul Parhi | EPFL Graph Search

Function-Space Optimality of Neural Architectures with Multivariate Nonlinearities

We investigate the function-space optimality (specifically, the Banach-space optimality) of a large class of shallow neural architectures with multivariate nonlinearities/activation functions. To that end, we construct a new family of Banach spaces defined ...

SIAM PUBLICATIONS2025

DISTRIBUTIONAL EXTENSION AND INVERTIBILITY OF THE k-PLANE TRANSFORM AND ITS DUAL

Michaël Unser, Rahul Parhi

We investigate the distributional extension of the k-plane transform in Rd and of related operators. We parameterize the k-plane domain as the Cartesian product of the Stiefel manifold of orthonormal k-frames in Rd with Rd-k. This parameterization imposes ...

2024

Modulation Spaces and the Curse of Dimensionality

Michaël Unser, Rahul Parhi

2023

The Sparsity of Cycle Spinning for Wavelet-Based Solutions of Linear Inverse Problems

Michaël Unser, Rahul Parhi

The usual explanation of the efficacy of wavelet-based methods hinges on the sparsity of many real-world objects in the wavelet domain. Yet, standard wavelet-shrinkage techniques for sparse reconstruction are not competitive in practice, one reason being t ...

IEEE Institute of Electrical and Electronics Engineers2023

Near-Minimax Optimal Estimation With Shallow ReLU Neural Networks

Rahul Parhi

We study the problem of estimating an unknown function from noisy data using shallow ReLU neural networks. The estimators we study minimize the sum of squared data-fitting errors plus a regularization term proportional to the squared Euclidean norm of the ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2023

Deep Learning Meets Sparse Regularization

Rahul Parhi

Deep learning (DL) has been wildly successful in practice, and most of the state-of-the-art machine learning methods are based on neural networks (NNs). Lacking, however, is a rigorous mathematical theory that adequately explains the amazing performance of ...

2023