Embedded Files
Welcome!
Di Wu, Ph.D.
Associate Professor @ Concordia, Department of Mechanical, Industrial and Aerospace Engineering
Associate Academic Member @ Mila (Quebec AI Institute)
Adjunct Professor @ McGill, Department of Electrical and Computer Engineering
About Me
About Me
Di Wu is an applied AI researcher working on Trustworthy and Data-Efficient Decision-Making for Complex Real-World Systems. He is currently an Associate Professor at Concordia University, an Associate Academic Member of Mila – Quebec AI Institute, and an Adjunct Professor at McGill University. Previously, he was a Senior Staff Research Scientist at Samsung AI Center Montreal, where he led the LLM Optimization team and the AI for Telecommunications team. Prior to Samsung, he conducted postdoctoral research at Mila and Stanford University. Di received his Ph.D. from McGill University (Montreal, Canada) in 2018 and his M.Sc. from Peking University (Beijing, China) in 2013. He also holds dual bachelor’s degrees in Microelectronics and Economics. His research focuses on enabling data-efficient, safe, and trustworthy decision-making and agentic AI systems for real-world problems under uncertainty, spanning three interconnected areas: Predictive Intelligence, Reliable (Safe) Decision-Making, and Trustworthy AI & Human-AI Collaboration. His work combines reinforcement learning, predictive modelling, time-series forecasting, optimization, and trustworthy generative AI to support operational intelligence in complex dynamic systems. Di works closely with real-world systems, including power systems, communication networks, transportation systems, and large-scale infrastructure. Multiple AI solutions developed by his team have been validated in real-world operational environments, demonstrating measurable improvements in energy efficiency, reliability, load balancing, peak reduction, and resource utilization. His long-term goal is to bridge advanced AI research with deployable and trustworthy decision-making systems for complex real-world operations. He has published over 200 papers, including work in leading AI conferences and journals such as NeurIPS, ICML, ICLR, AAAI, IJCAI, Journal of Machine Learning Research, Transactions on Machine Learning Research, IEEE Transactions on AI, IEEE Transactions on Smart Grid, IEEE Transactions on Intelligent Transportation Systems, and IEEE Journal on Selected Areas in Communications. He has secured over CAD $3 million in competitive and partnered research funding from funding agencies and industry partners. Di also holds more than 20 U.S. patents related to reinforcement learning, forecasting, trustworthy AI, large language models, load balancing, energy management, anomaly detection, and large-scale system optimization, including multiple patents recognized as strategically important. His work has received three Best Paper Awards—IEEE GLOBECOM 2021 (AI for Telecommunications), IEEE ITSC 2024 (AI for Transportation), and IEEE EPEC 2025 (AI for Energy)—as well as multiple Best Paper Award nominations.
Di Wu is an applied AI researcher working on Trustworthy and Data-Efficient Decision-Making for Complex Real-World Systems. He is currently an Associate Professor at Concordia University, an Associate Academic Member of Mila – Quebec AI Institute, and an Adjunct Professor at McGill University. Previously, he was a Senior Staff Research Scientist at Samsung AI Center Montreal, where he led the LLM Optimization team and the AI for Telecommunications team. Prior to Samsung, he conducted postdoctoral research at Mila and Stanford University. Di received his Ph.D. from McGill University (Montreal, Canada) in 2018 and his M.Sc. from Peking University (Beijing, China) in 2013. He also holds dual bachelor’s degrees in Microelectronics and Economics. His research focuses on enabling data-efficient, safe, and trustworthy decision-making and agentic AI systems for real-world problems under uncertainty, spanning three interconnected areas: Predictive Intelligence, Reliable (Safe) Decision-Making, and Trustworthy AI & Human-AI Collaboration. His work combines reinforcement learning, predictive modelling, time-series forecasting, optimization, and trustworthy generative AI to support operational intelligence in complex dynamic systems. Di works closely with real-world systems, including power systems, communication networks, transportation systems, and large-scale infrastructure. Multiple AI solutions developed by his team have been validated in real-world operational environments, demonstrating measurable improvements in energy efficiency, reliability, load balancing, peak reduction, and resource utilization. His long-term goal is to bridge advanced AI research with deployable and trustworthy decision-making systems for complex real-world operations. He has published over 200 papers, including work in leading AI conferences and journals such as NeurIPS, ICML, ICLR, AAAI, IJCAI, Journal of Machine Learning Research, Transactions on Machine Learning Research, IEEE Transactions on AI, IEEE Transactions on Smart Grid, IEEE Transactions on Intelligent Transportation Systems, and IEEE Journal on Selected Areas in Communications. He has secured over CAD $3 million in competitive and partnered research funding from funding agencies and industry partners. Di also holds more than 20 U.S. patents related to reinforcement learning, forecasting, trustworthy AI, large language models, load balancing, energy management, anomaly detection, and large-scale system optimization, including multiple patents recognized as strategically important. His work has received three Best Paper Awards—IEEE GLOBECOM 2021 (AI for Telecommunications), IEEE ITSC 2024 (AI for Transportation), and IEEE EPEC 2025 (AI for Energy)—as well as multiple Best Paper Award nominations.
Page updated
Google Sites
Report abuse