معرفی
Garrett Ethan Katz is an Associate Professor in the Department of Electrical Engineering & Computer Science at Syracuse University. His research focuses on vertically integrated artificial intelligence, spanning low-level robotic control to high-level reasoning. He is actively involved in projects such as single-pass learning, automated algorithm discovery, and neuro-symbolic robotic control. His lab explores the intersection of cognitive modeling, machine learning, and robotics.
Education:
- B.A. Philosophy, Cornell University, 2007
- M.A. Mathematics, City College of New York, 2011
- Ph.D. Computer Science, University of Maryland, College Park, 2017
Research Interests:
- Single-pass learning rules with high capacity and efficiency
- Robotic imitation learning using cause-effect reasoning frameworks (CERIL)
- Neuro-symbolic programming and neural virtual machines
- Numerical analysis of neural network dynamics
- Human-robot teaming environments and cognitive workload assessment
Publications Overview: Recent work emphasizes theoretical impossibility results in single-pass learning, frameworks for robotic imitation learning, and neurocomputational models that bridge symbolic algorithms and neural networks. His research also explores applications in cryo-electron microscopy and human cognitive-motor performance analysis.
Awards:
- Best Paper Award (Augmented Cognition at HCII, 2024)
- SAI Computing Conference Best Paper Award (2020)
- Larry S. Davis Doctoral Dissertation Award (2018)
- Best Student Paper (Artificial General Intelligence Conference, 2016)
Advising & Teaching: Currently advising PhD students Naveed Tahir, Akshay, Xulin Chen, Borui He, and Ruipeng Liu. Former advisees include Khushboo Gupta (M.S. 2019) and Chad Thom Smith (Summer REU 2023). Teaches courses like CIS 700 (Advanced Topics in AI), CIS 467/667 (Intro to AI), and ECS102/CIS151 (Programming Fundamentals).
Labs & Projects: Leads research on neurocognitive robotic systems using programmable neural architectures. Active in developing frameworks like CERIL and NeuroCERIL for humanoid robot learning. Collaborates on projects applying neural virtual machines for algorithmic tasks and motor control.
