Machine Learning Assisted Improvements to Multi-Criterion Optimization AlgorithmsSpeaker: Kalyanmoy Deb – MI, United States
Topic(s): Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing , Software Engineering and Programming , Computational Theory, Algorithms and Mathematics , Information Systems, Search, Information Retrieval, Database Systems, Data Mining, Data Science , Applied Computing
Multi-criterion optimization problems give rise to a set of Pareto-optimal solutions, which first must be found before a single preferred solution is chosen for implementation. To find a number of trade-off solutions, a population-based optimization algorithm, such as the evolutionary multi-objective optimization (EMO) method, is called for. An EMO algorithm involves a number of operations – initialization of a population of solutions, definition of domination, surrogate function development, a selection operator, a set of creation operators, and termination criterion. Each of these operations can be potentially improved with a suitable machine learning method by using data collected from the previous iterations within an EMO run. The learnt machine learning models can then be used to improve the performance of EMO algorithms, In this talk, we shall discuss some of these implementations and their benefits in arriving at a set of Pareto-optimal solutions faster and with more reliability. Examples from test problems and real-world problems will be presented. The lecture will motivate participants to go beyond each of machine learning and EMO fields and provide ideas of combining their strengths to develop better multi-criterion optimization algorithms.
The participants will get a good knowledge of EMO algorithms and also key machine learning methods. Different hybridizations of ML methods to EMO will provide participants new ways to combine the two fields together.
About this LectureNumber of Slides: 50
Duration: 50 minutes
Languages Available: English
Request this Lecture
To request this particular lecture, please complete this online form.
Request a Tour
To request a tour with this speaker, please complete this online form.
All requests will be sent to ACM headquarters for review.