Delfi: Online Planner Selection for Cost-Optimal Planning

Page Contributors: Adam Green,

tags: IPC2018 Optimising

Planner Quality: -

Year Published: 2018

Paper: Delfi: Online Planner Selection for Cost-Optimal Planning [ Katz, M. Sohrabi, S. Horst Samulowitz, H. Sievers, S. ]

Preceded By: Fast Downward [Link Needed]

Delfi is an online portfolio planner. In contrast to existing techniques, Delfi exploits deep learning techniques to learn a model that predicts which of the planners in the portfolio can solve a given planning task within the imposed time and memory bounds. Delfi uses graphical representations of a planning task which allows exploiting existing tools for image convolution

Support

Delfi has not been tested with eviscerator yet. Delfi is designed for classical domains and so most likely does not support temporal numeric planning.

Downloading and Compiling Delfi

Source code for Delfi1 can be found on a BitBucket Submission to IPC 2018

Source code for Delfi2 can be found on a BitBucket Submission to IPC 2018