Published February 17, 2020
| Version v1
Software
Open
Code from the paper "Neural Network Heuristics for Classical Planning: A Study of Hyperparameter Space"
Authors/Creators
- 1. University of Basel
- 2. Saarland University
Description
This repository contains all benchmarks, test instances, and the source code for the publication:
Patrick Ferber, Malte Helmert, and Jörg Hoffmann. "Neural Network Heuristics for Classical Planning: A Study of Hyperparameter Space". 24th European Conference on Artificial Intelligence (ECAI). 2020
The paper can be found here.
The trained models of the competitive performance can be found here.
Notes
Patrick Ferber was funded by DFG grant 389792660 as part of TRR 248 (see https://perspicuous-computing.science). This work was supported by the Swiss National Science Foundation (SNSF) as part of the project Certified Correctness and Guaranteed Performance for Domain-IndependentPlanning (CCGP-Plan).
Files
ferber-et-al-ecai2020.zip
Files
(56.0 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:3fb3a2ba3539912ba3ef3a79fad2e6a1
|
56.0 MB | Preview Download |
Additional details
Funding
- Swiss National Science Foundation
- Certified Correctness and Guaranteed Performance for Domain-Independent Planning (CCGP-Plan) 200021_182107