1 code implementation • 15 Jan 2021 • Miquel Junyent, Vicenç Gómez, Anders Jonsson
In classical planning, we show how IW(1) at two levels of abstraction can solve problems of width 2.
1 code implementation • 12 Apr 2019 • Miquel Junyent, Anders Jonsson, Vicenç Gómez
Surprisingly, we observe that the representation learned by the neural network can be used as a feature space for the width-based planner without degrading its performance, thus removing the requirement of pre-defined features for the planner.
no code implementations • 15 Jun 2018 • Miquel Junyent, Anders Jonsson, Vicenç Gómez
The planning step hinges on the Iterated-Width (IW) planner, a state of the art planner that makes explicit use of the state representation to perform structured exploration.