Roby Velez, Ph.D.
Assistant Professor
- Milwaukee WI UNITED STATES
- Diercks Hall DH422
- Electrical Engineering and Computer Science
Dr. Roby Velez is an expert in artificial intelligence research involving the analysis of Artificial Neural Networks.
Education, Licensure and Certification
Ph.D.
Computer Science
University of Wyoming
2019
M.S.
Evolutionary and Adaptive Systems
University of Sussex
2012
B.S.
Engineering
Swarthmore College
2009
Biography
development of learning algorithms. He is also interested in STEM outreach and teaching.
Industry Expertise
Areas of Expertise
Social
Media Appearances
UW’s Jeff Clune Receives Prestigious NSF Award to Evolve Artificially Intelligent Robots
UW News online
2015-09-08
Clune’s Ph.D. student Roby Velez has volunteered long hours to help get the club off the ground, and this grant will provide funding for Velez and other graduate students to be able to dedicate more time to refining the club’s learning materials.
How robots learn general skills
Phys.org online
2014-01-08
To understand ourselves better, Roby Velez researches how robots learn general skills that help them explore their environment.
Research Grants
Active Learning Initiatives
UW Tier-1 Engineering Initiative
2015
Engineering’s Next Generation Program
Ellbogen Foundation at the University of Wyoming
2015-2019
Selected Publications
Diffusion-based neuromodulation can eliminate catastrophic forgetting in simple neural networks
PloS one2017
A long-term goal of AI is to produce agents that can learn a diversity of skills throughout their lifetimes and continuously improve those skills via experience. A longstanding obstacle towards that goal is catastrophic forgetting, which is when learning new information erases previously learned information. Catastrophic forgetting occurs in artificial neural networks (ANNs), which have fueled most recent advances in AI.
Identifying Core Functional Networks and Functional Modules within Artificial Neural Networks via Subsets Regression
GECCO '16: Proceedings of the Genetic and Evolutionary Computation Conference 20162016
As the power and capabilities of Artificial Neural Networks (ANNs) grow, so do their size and complexity. To both decipher and improve ANNs, we need to build better tools that help us understand their inner workings. To that end, we introduce an algorithm called Subsets Regression on network Connectivity (SRC). SRC allows us to prune away unimportant nodes and connections in ANNs, revealing a core functional network (CFN) that is simpler and thus easier to analyze.
Novelty search creates robots with general skills for exploration
GECCO '14: Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation2014
Novelty Search, a new type of Evolutionary Algorithm, has shown much promise in the last few years. Instead of selecting for phenotypes that are closer to an objective, Novelty Search assigns rewards based on how different the phenotypes are from those already generated. A common criticism of Novelty Search is that it is effectively random or exhaustive search because it tries solutions in an unordered manner until a correct one is found.