Scott Sanner

Scott Sanner is a Professor of Industrial Engineering at the University of Toronto, cross-appointed in Computer Science and a faculty affiliate at the Vector Institute. He holds a PhD from the University of Toronto (2008), an MS from Stanford University (2002), and a double BS from Carnegie Mellon University (1999). His research bridges data-driven fields (Machine Learning, Information Retrieval) with decision-driven fields (Artificial Intelligence, Operations Research), focusing on applications like conversational recommenders, smart cities, and adaptive systems.

Professor Sanner’s work addresses challenges in traffic optimization, power systems security, and healthcare analytics. He has pioneered methods like the eMARLIN traffic control framework and contributed to datasets like CoLoTa for commonsense reasoning. His academic service includes co-chairing ICAPS 2023 and editorial roles at AIJ, JAIR, and MLJ.

Key achievements include the Google Faculty Research Award (2020) and paper awards from the AI Journal (2014), Transport Research Board (2016), and CPAIOR (2018). His lab, the Data-Driven Decision Making (D3M) group, explores intersections of AI, optimization, and real-world systems.

• Education:
  • PhD in Computer Science, University of Toronto (2008)
  • MS in Computer Science, Stanford University (2002)
  • BS in Computer Science & Electrical Engineering, Carnegie Mellon University (1999)
• Awards:
  • AI Journal Paper Award (2014)
  • Transport Research Board Paper Award (2016)
  • CPAIOR Paper Award (2018)
  • Google Faculty Research Award (2020)
• Grants & Roles:
  • Dean’s Spark Professorship (2018-2021)
  • Principal Investigator on multiple traffic signal control projects

His research trends emphasize generative models in recommendations, verifiable commonsense reasoning, and scalable traffic optimization using reinforcement learning. Recent work explores LLM-driven systems and ethical evaluation frameworks for conversational agents.