Mauricio Humberto Ferrato

Computer Science PhD Candidate at the University of Delaware.

Download CV

About

Mauricio

Mauricio is a Computer Science PhD Candidate currently enrolled at the University of Delaware.

His areas of interest involve high performance computing, focusing on porting computational science applications to run on accelerators using high-level parallel programming models and developing and scaling deep learning models on supercomputers.

Mauricio is a member of the CRPL group at the University of Delaware and is advised by Dr. Sunita Chandrasekaran.

Education

September 2018 - Present

University of Delaware (Newark, Delaware)

PhD in Computer Science

September 2013 - May 2018

University of Delaware (Newark, Delaware)

Bachelor of Science in Computer Engineering

Minor in Computer Science

Experience

September 2018 - Present

Research Assistant

University of Delaware (Newark, Delaware)

Working with Nemours/Alfred I. duPont Hospital for Children using machine learning techniques to develop different prediction models for oncology and sickle cell patients using EHR and genomic sequence data.

Helping build the MPI ACCEL benchmark suite in collaboration with the Standard Performance Evaluation Corporation High Performance Group (SPEC HPG)

June 2018 - August 2018

Summer Intern

National Cancer Institute (NCI) (Rockville, Maryland)

Worked as a summer intern for the National Cancer Institute (NCI) on various Pilot 1 deep learning models of the ECP-CANDLE project showing scalability of multi-GPU training using Horovod and finding uncertainty quantification using dropout at inference as a Bayesian approximation.

January 2017 - May 2018

Undergraduate Project

University of Delaware (Newark, Delaware)

In collaboration with Department of Chemistry & Biology’s Prof. Juan Perilla and under Prof. Sunita Chandrasekaran’s advisement, accelerated a chemical shift prediction algorithm that provides insight into protein secondary structure using the high-level parallel programming model OpenACC.

Journal publication about this project currently under review (Parallelizing Chemical Shift with Portable Programming Model. Cell.)

Research area/Skills

GPUs/Accelerators

In the era of accelerators, transitioning large scale applications to heterogenous systems that take advantage of the host-device execution model is the path to improve and optimize runtime performance looking ahead on the future of exascale computing.

Deep Learning

The world of machine learning is thriving due to the expanding growth of big data. More information can be gathered, and better accuracy can be found when a model is trained over large amounts of data. However, more data means the process can become quite expensive. For this reason, computer scientists must begin looking at HPC techniques to efficiently reduce maximize performance while reducing cost.

Programming

C/C++

Python

Java

Tensorflow/Keras

OpenACC

OpenMP

MPI

CUDA

Contact