Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Hrayr Harutyunyan

Estimating informativeness of samples with Smooth Unique Information


Jan 17, 2021
Hrayr Harutyunyan, Alessandro Achille, Giovanni Paolini, Orchid Majumder, Avinash Ravichandran, Rahul Bhotika, Stefano Soatto

* ICLR 2021, 22 pages 

  Access Paper or Ask Questions

Improving Generalization by Controlling Label-Noise Information in Neural Network Weights


Feb 19, 2020
Hrayr Harutyunyan, Kyle Reing, Greg Ver Steeg, Aram Galstyan


  Access Paper or Ask Questions

Efficient Covariance Estimation from Temporal Data


May 30, 2019
Hrayr Harutyunyan, Daniel Moyer, Hrant Khachatrian, Greg Ver Steeg, Aram Galstyan


  Access Paper or Ask Questions

MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing


May 28, 2019
Sami Abu-El-Haija, Bryan Perozzi, Amol Kapoor, Nazanin Alipourfard, Kristina Lerman, Hrayr Harutyunyan, Greg Ver Steeg, Aram Galstyan


  Access Paper or Ask Questions

MixHop: Higher-Order Graph Convolution Architectures via Sparsified Neighborhood Mixing


Apr 30, 2019
Sami Abu-El-Haija, Bryan Perozzi, Amol Kapoor, Hrayr Harutyunyan, Nazanin Alipourfard, Kristina Lerman, Greg Ver Steeg, Aram Galstyan

* International Conference on Machine Learning, 2019 

  Access Paper or Ask Questions

Disentangled Representations via Synergy Minimization


Oct 10, 2017
Greg Ver Steeg, Rob Brekelmans, Hrayr Harutyunyan, Aram Galstyan

* 8 pages, 4 figures, 55th Annual Allerton Conference on Communication, Control, and Computing, 2017 

  Access Paper or Ask Questions

Multitask Learning and Benchmarking with Clinical Time Series Data


Mar 22, 2017
Hrayr Harutyunyan, Hrant Khachatrian, David C. Kale, Aram Galstyan

* Proposes clinical prediction benchmark tasks; details can be found at https://github.com/turambar/mimic3-benchmarks. Manuscript revisions will be uploaded regularly to reflect updates to the benchmark data set, new results, and new findings. An earlier version of this manuscript is currently under review for SIGKDD 2017 

  Access Paper or Ask Questions