Evolutionary Machine Learning in Engineering Design - Day 4 Morning

Explore evolutionary machine learning techniques for automated engineering design in this comprehensive tutorial lecture from JSALT 2025. Learn how evolutionary algorithms can revolutionize the design of programs, electronic circuits, neural networks, antennas, and other engineering objects through holistic, multi-objective approaches that create solutions with unique properties. Discover the key ingredients of evolutionary design methods with a focus on genetic programming, and examine real-world examples of evolved solutions including approximate arithmetic circuits, neural network architectures, and image filters that demonstrate superior properties compared to conventional designs. Understand how evolutionary computing techniques enrich machine learning methods and vice versa, creating powerful hybrid approaches for complex engineering challenges. Gain insights from Professor Lukáš Sekanina of Brno University of Technology, a leading expert in genetic programming, approximate computing, and evolvable hardware who has contributed over 250 research papers to the field and served as principal investigator on multiple Czech Science Foundation projects.