Decision Focused Learning for Prediction+ Optimisation Problems

  • Maxime Albert Mulamba Ke Tchomba (Speaker)
  • Emilio Gamba (Speaker)
  • Tias Guns (Speaker)

Activity: Talk or presentation › Talk or presentation at a workshop/seminar

Description

Increasingly, constraint programming problems can not be fully specified as facts, but part of the problem input is predicted using machine learning; for example demand and price in energy scheduling, (truck)load in vehicle transport, or even visual or natural language input. This talk has two purposes. First, we will highlight two settings in which a tighter integration between the machine learning and the model-then-solve leads to improved results. Next, we introduce the CPMpy constraint programming library, which was made to be highly compatible with machine learning libraries, in order to use the output of one in the other and reversely
Period7 Feb 2023
Event titleAAAI 2023 Bridge on Constraint Programming and Machine Learning
Event typeWorkshop
Degree of RecognitionInternational