The TILOS AI Institute: Integrating optimization and AI for chip design, networks, and robotics

Abstract

Optimization is a universal quest, reflecting the basic human need to do better. Improved optimizations of energy-efficiency, safety, robustness, and other criteria in engineered systems would bring incalculable societal benefits. But, fundamental challenges of scale and complexity keep many such real-world optimization needs beyond reach. This article describes The Institute for Learning-enabled Optimization at Scale (TILOS), an NSF AI Research Institute for Advances in Optimization that aims to overcome these challenges in three high-stakes use domains: chip design, communication networks, and contextual robotics. TILOS integrates foundational research, translation, education, and broader impacts toward a new nexus of optimization, AI, and data-driven learning. We summarize central challenges, early progress, and futures for the institute.