Mojtaba Sahraee Ardakan — accepted AI-conference papers

2020

Matrix Inference and Estimation in Multi-Layer Models

Parthe Pandit, Mojtaba Sahraee Ardakan, Sundeep Rangan, Philip Schniter, Alyson K. Fletcher

NeurIPS 2020poster

We consider the problem of estimating the input and hidden variables of a stochastic multi-layer neural network from an observation of the output. The hidden variables in each layer are represented as matrices with statistical interactions along both rows as well as columns. This problem applies to…

Cited by 13 Source ↗PDF ↗code ↗Scholar ↗

2019

Input-Output Equivalence of Unitary and Contractive RNNs

Melikasadat Emami, Mojtaba Sahraee Ardakan, Sundeep Rangan, Alyson K. Fletcher

NeurIPS 2019poster

Unitary recurrent neural networks (URNNs) have been proposed as a method to overcome the vanishing and exploding gradient problem in modeling data with long-term dependencies. A basic question is how restrictive is the unitary constraint on the possible input-output mappings of such a network? This…

Cited by 5 Source ↗PDF ↗code ↗Scholar ↗