MIT Laboratory for Information and Decision Systems (LIDS)

A faster problem-solving tool that guarantees feasibility The FSNet system, developed by LIDS grad student Hoang Nguyen and LIDS PI Priya Donti, could help power grid operators rapidly find feasible solutions for optimizing the flow of electricity. This new problem-solving tool incorporates a feasibility-seeking step into a powerful machine-learning model trained to solve the problem. The feasibility-seeking step uses the model’s prediction as a starting point, iteratively refining the solution until it finds the best achievable answer. In addition to helping schedule power production in an electric grid, this new tool could be applied to many types of complicated problems, such as designing new products, managing investment portfolios, or planning production to meet consumer demand. Learn more and read the paper at MIT News: https://bit.ly/3JzogJh MIT EECS MIT Schwarzman College of Computing MIT School of Engineering

👀 Youssef Marzouk was recently appointed associate dean of the MIT Schwarzman College of Computing! The LIDS PI and MIT AeroAstro Professor will work in his new role to foster a stronger community among bilingual computing faculty across MIT. A key aspect of this work will be providing additional structure and support for faculty members who have been hired into shared positions in departments and the college. Learn more at MIT News: https://bit.ly/4oRwlbf MIT School of Engineering

And the winner is… 🎉 LIDS alum Jason Altschuler SM ’18, PhD ’22 and PI Pablo Parrilo have won the 2025 INFORMS Computing Society (ICS) Prize for outstanding contributions at the intersection of computing and operations research. Their pioneering work on accelerating gradient descent through stepsize hedging offers powerful new tools for optimization. Learn more and read their award-winning papers: ➡️ “Acceleration by Stepsize Hedging: Multi-Step Descent and the Silver Stepsize Schedule” ➡️ “Acceleration by Stepsize Hedging: Silver Stepsize Schedule for Smooth Convex Optimization” 🔗 http://bit.ly/4o7dsAU MIT EECS MIT Department of Mathematics MIT Schwarzman College of Computing MIT School of Engineering The Wharton School

🚀 𝐀𝐫𝐞 𝐲𝐨𝐮 𝐢𝐧 𝐁𝐨𝐬𝐭𝐨𝐧 𝐧𝐞𝐱𝐭 𝐌𝐨𝐧𝐝𝐚𝐲? 𝐃𝐨𝐧’𝐭 𝐦𝐢𝐬𝐬 𝐭𝐡𝐢𝐬! I’m thrilled to host my friend Aaron Ames (Caltech) at MIT for the MIT Laboratory for Information and Decision Systems (LIDS) 𝐒𝐞𝐦𝐢𝐧𝐚𝐫 on 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧𝐬 𝐟𝐨𝐫 𝐒𝐚𝐟𝐞 𝐀𝐮𝐭𝐨𝐧𝐨𝐦𝐲: 𝐖𝐡𝐲 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐍𝐞𝐞𝐝𝐬 𝐂𝐨𝐧𝐭𝐫𝐨𝐥 🗓️ 𝐖𝐡𝐞𝐧: 𝐌𝐨𝐧𝐝𝐚𝐲, 𝐎𝐜𝐭𝐨𝐛𝐞𝐫 𝟐𝟎, 𝟐𝟎𝟐𝟓 — 𝟒:𝟎𝟎 𝐏𝐌 📍 𝐖𝐡𝐞𝐫𝐞: 𝐑𝐨𝐨𝐦 𝟒𝟓–𝟐𝟑𝟎 — 𝐌𝐈𝐓 𝐒𝐜𝐡𝐰𝐚𝐫𝐳𝐦𝐚𝐧 𝐂𝐨𝐥𝐥𝐞𝐠𝐞 𝐨𝐟 𝐂𝐨𝐦𝐩𝐮𝐭𝐢𝐧𝐠 If you care about the future of autonomy, robotics, and control-aware learning, this is the place to be. https://lnkd.in/dqjmMUTf #MIT #LIDS #Seminar #Autonomy #Control #Learning #Robotics #AI #SafeAutonomy

MIT researchers are making fusion energy safer and more reliable 🔥⚡️ LIDS PI Chuchu Fan, grad student Oswin So, and colleagues have created a prediction model that combines physics and machine learning to prevent damaging disruptions when tokamak reactors power down — an advance that could boost the reliability of future fusion power plants. Learn more at MIT News and read the paper in Nature Communications: https://bit.ly/4nZyv8H MIT contributors include: Cristina Rea, Allen M. Wang, Oswin So, Charles Dawson, and Chuchu Fan Swiss Plasma Center at EPFL contributors include: Alessandro Pau, Olivier Sauter, Anna Vu, Cristian Galperti, Antoine Merle, Yoeri Poels, Cristina Venturini, Federico Felici & Stefano Marchioni Credits: Thanks to Mark D. Boyer, Commonwealth Fusion Systems, Devens, MA, USA and to the TCV team for enabling this work. MIT AeroAstro Plasma Science and Fusion Center at MIT MIT Schwarzman College of Computing MIT School of Engineering

🌱🤖 Fighting for the health of the planet with #AI. Priya Donti is applying machine learning to optimize renewable energy. “Machine learning is already really widely used for things like solar power forecasting, which is a prerequisite to managing and balancing power grids,” says the MIT EECS assistant professor and LIDS PI. “My focus is: How do you improve the algorithms for actually balancing power grids in the face of a range of time-varying renewables?” Read the full profile on MIT News: https://bit.ly/4o6Mnxw 💡🔌 Climate Change AI MIT Schwarzman College of Computing

Designing for uncertainty A new framework from LIDS PI Gioele Zardini, LIDS grad student Yujun Huang, and University of Zurich’s Marius Furter can help engineers design complex systems that involve many interconnected parts, such as delivery drones that navigate changing environments, in a way that explicitly accounts for the uncertainty in each component’s performance. The approach could enable autonomous vehicles, commercial aircraft, or transportation networks that are more reliable in the face of real-world unpredictability. More: http://bit.ly/48dsBf9 MIT Civil and Environmental Engineering MIT Institute for Data, Systems, and Society (IDSS) MIT School of Engineering MIT Schwarzman College of Computing

The pros and cons of synthetic data in #AI. Synthetic data — artificial data built by algorithms — are gaining traction in AI. They can slash costs, speed up model training, and protect privacy, but their limitations require careful planning and evaluation. Read a 3Q with LIDS PRS and DataCebo CEO Kalyan Veeramachaneni to learn more. 🧠💡 https://bit.ly/3IlYnMu MIT EECS MIT School of Engineering MIT Schwarzman College of Computing

Thrilled to share details about the new MEng track in Data Science for Engineering Systems (DSES) at Massachusetts Institute of Technology, which I’m honored to help lead. This program equips students with cutting-edge data science and computational modeling skills to tackle some of society’s biggest challenges: sustainable energy, resilient infrastructure, circular materials, adaptive supply chains, and future-ready urban systems. 🌍⚡🏙️ Curious to learn more? Join our Admissions Information Session to explore how this program prepares the next generation of engineers to lead in sustainability and resilience. 📅 [Link to register] 👉 https://lnkd.in/d7gsSjEm Let’s shape the future of engineering systems together! MIT Civil and Environmental Engineering MIT Laboratory for Information and Decision Systems (LIDS) MIT School of Engineering MIT Institute for Data, Systems, and Society (IDSS)