News

Feb 2019: Invited talk at the SIAM Conference on Computational Science and Engineering as part of the Minisymposium on High Performance Sparse Matrix, Tensor, and Graph Kernels.
Dec 2018: I was featured in the annual report of the School of Computational Science & Engineering at Georgia Tech.
Dec 2018: Attending NeurIPS 2018 in Montreal, one workshop paper accepted at the Machine Learning for Health Workshop.
October 2018: Successfully defended my thesis proposal!
Aug 2018: Attending KDD 2018 in London to present our work on integer tensor factorization applied to computational phenotyping.
July 2018: Invited talk at Microsoft Research Cambridge as part of the Health Hive seminar.
June 2018: Invited talk at the University of Sheffield as part of the Machine Learning seminar series.
May 2018: Invited talk at the SIAM Conference on Applied Linear Algebra in Hong Kong as part of the Minisymposium on Constrained Low-Rank Matrix and Tensor Approximations.

 

Publications


Temporal Phenotyping of Medically Complex Children via PARAFAC2 Tensor Factorization
Journal of Biomedical Informatics

Paper link


SUSTain: Scalable Unsupervised Scoring for Tensors and its Application to Phenotyping
KDD 2018 [Oral Presentation (Acceptance Rate: 10.9%)]

PDF Code Promotional Video


COPA: Constrained PARAFAC2 for Sparse & Large Datasets
CIKM 2018

PDF Arxiv version


SPARTan: Scalable PARAFAC2 for Large & Sparse Data
KDD 2017 [Oral Presentation (Acceptance Rate: 8.6%)]

PDF Code Slides Poster Promotional Video Video of Oral Presentation


Polyadic Regression and its Application to Chemogenomics
SDM 2017

PDF Slides Supplementary Material Poster


Model-Driven Sparse CP Decomposition for High-Order Tensors
IPDPS 17

PDF Jiajia's slides


CP-ORTHO: An Orthogonal Tensor Factorization Framework for Spatio-Temporal Data
SIGSPATIAL 2017

PDF


SPALS: Fast Alternating Least Squares via Implicit Leverage Scores Sampling
NeurIPS 2016

PDF Dehua's code


Sparse Hierarchical Tucker Factorization and its Application to Healthcare
ICDM 2015

PDF Code Slides Extended version


An input-adaptive and in-place approach to dense tensor-times-matrix multiply
SuperComputing 15

PDF Jiajia's Slides

Contact

  • perros@gatech.edu
  • Klaus Advanced Computing Building, 266 Ferst Dr, Atlanta, GA 30332