Evolutionary Computation and Decision Making

Evolutionary Computation and Decision Making

EC + MCDM 2022

Keynote speaker:  Prof. Julia Handl

Title: Evolutionary Computation and Decision Making in Unsupervised Learning

Abstract: Despite advances in AutoML, humans, and domain expertise held by humans, continue to be instrumental to success in most machine learning applications. The development of machine learning pipelines is typically approached in a highly iterative fashion, highlighting the importance of repeated human input in arriving at model designs that help capture the essence, and practicalities, of a given learning problem. This talk will discuss the potential of evolutionary computation in supporting decision making for unsupervised learning problems, considering opportunities for both automated and interactive decisions.

Accepted papers

  1. Aghaei Pour, Bandaru, Afsar, Miettinen. Desirable Properties in Performance Indicators for Interactive Evolutionary Methods
  2. Chugh. R-MBO: A Multi-surrogate Approach for Preference Incorporation in Multi-objective Bayesian Optimisation
  3. Larraga Maldonado, Miettinen. Interactive MOEA/D with Multiple Types of Preference Information
  4. Almeida, Lezama, Soares, Vale, Canizes. Preliminary Results of Advanced Heuristic Optimization in the Risk-based Energy Scheduling Competition
  5. Mazumdar, Otayagich, Miettinen. Interactive Evolutionary Multiobjective Optimization with Modular Physical User Interface