Project Grant 2443024

This Project Grant award of $342,235, provided by the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Federal Grant Program (CFDA 47.070), aims to develop novel explainability tools and techniques to enable effective human-AI collaboration during the design and deployment of learning-based network controllers.

The key objectives of this 5-year project at the University of California, Irvine are to: 1) develop a concept-based explainer that interprets model behavior using high-level, human-understandable concepts; 2) create a hybrid explainability framework that integrates concept-based, low-level feature-based, and predictive future performance-based explanations; and 3) investigate how these explanations can enhance operational tasks like data curation, test coverage evaluation, debugging, and root cause analysis. This research seeks to bridge the gap between black-box machine learning models and practical network management needs, enabling network operators to better understand, trust, and manage learning-based controllers in real-world network environments.