David DeFazio

I am a Ph.D. student in the Department of Computer Science, SUNY Binghamton where I work with Prof. Shiqi Zhang. My work focuses on the application of reinforcement learning for robotic tasks. In particular, I'm studying how we can best leverage human knowledge to more efficiently solve sparse reward tasks. I am part of the Autonomous Intelligent Robotics Group (AIR).

Interests

Application of Reinforcement Learning to Robotics

Knowledge based Reinforcement Learning

Graph Neural Networks

Publications

David DeFazio and Arti Ramesh, "Adversarial Model Extraction on Graph Neural Networks", AAAI Workshop on Deep Learning on Graphs: Methodologies and Applications (DLGMA), 2020 [Pdf]
David DeFazio, Arti Ramesh, and Anand Seetharam, "NYCER: A Non-Emergency Response Predictor for NYC using Sparse Gaussian Conditional Random Fields", International Conference on Mobile and Ubiquitous Systems (MobiQuitous), 2018 [Pdf]
Gissella Bejarano, David DeFazio, and Arti Ramesh, "Deep Latent Generative Models For Energy Disaggregation", AAAI Conference on Artificial Intelligence, 2019. [Pdf]

Education

2019 - Present

PhD in Computer Science

SUNY Binghamton

2018 - 2019

MS in Computer Science

SUNY Binghamton

2014 - 2018

BS in Computer Science

SUNY Binghamton

Experience

June 2020 - August 2020

Air Force Research Laboratory Scholars Program

Explored connection between Meta-Learners and generalization of neural networks

Fall 2018 - Present

Graduate Teaching Assistant

Automata Theory, Object Oriented Programming, Architechture from a Programming Perspective

Contact

Address
N-09 Engineering Building
Department of Computer Science
SUNY Binghamton
NY 13902

Email
ddefazi1@binghamton.edu