Deep neural networks and other modern machine learning techniques are revolutionizing various fields, such as transportation and healthcare, by uncovering patterns in large data sets and making accurate predictions. Deep generative models, in particular, have shown great potential in generating new design ideas, which is crucial for innovation and the development of new products. These automated methods can assist human experts, who often rely on experience and heuristics to explore design ideas through a time-consuming iterative process. When combined with optimization methods and physics-based simulators, deep generative models can synthesize high-performance realistic designs. In this talk, we will examine the main challenges deep generative models face in design problems and how novel methods that model performance, constraints, and novelty, address these challenges. Finally, we will discuss the broader applications of these data-driven design methods to applications such as aerodynamic design, kinematic design, and topology optimization.
Faez Ahmed is the d'Arbeloff career development assistant professor in the Department of Mechanical Engineering at MIT, where he leads the Design Computation and Digital Engineering (DeCoDE) lab. His research focuses on developing new machine learning and optimization methods to study complex engineering design problems. Ahmed and his team are interested in a variety of questions related to the design of complex systems, including how algorithms can synthesize high-performing designs that meet real-world requirements, how algorithms can help discover or create creative designs that have never been seen before, and how distributed teams of people can work together to create better products. Before joining MIT, Ahmed was a postdoctoral fellow at Northwestern University and completed his Ph.D. in mechanical engineering at the University of Maryland. He also worked in the railway and mining industry in Australia, where he pioneered data-driven predictive maintenance and renewal planning efforts.