Software and Data

Here is a list of software and data resources that we developed as part of the RLeap project. You can find additional repositories on our GitHub page.

Libraries

Mimir: A Generalized Planning Library

Description Logics State Features for Planning Library (DLPlan)

Softwares

Learning General Policies From Examples

Learning Lifted STRIPS Models from Action Traces Alone: A Simple, General, and Scalable Solution

Learning More Expressive General Policies for Classical Planning Domains

Learning to Ground Existentially Quantified Goals

Symmetries and Expressive Requirements for Learning General Policies

Expressing and Exploiting Subgoal Structure in Classical Planning Using Sketches

On Policy Reuse: An Expressive Language for Representing and Executing General Policies that Call Other Policies

Learning generalized policies for fully observable non-deterministic planning domains

General and Reusable Indexical Policies and Sketches

Learning General Policies with Policy Gradient Methods

Learning Hierarchical Policies by Iteratively Reducing the Width of Sketch Rules

Learning Generalized Policies without Supervision Using GNNs

Learning Sketches for Decomposing Planning Problems into Subproblems of Bounded Width: Extended Version

Learning General Optimal Policies with Graph Neural Networks: Expressive Power, Transparency, and Limits

Flexible FOND Planning with Explicit Fairness Assumptions

Learning General Planning Policies from Small Examples Without Supervision

Learning First-Order Representations for Planning from Black Box States: New Results

Learning First-Order Symbolic Representations for Planning from the Structure of the State Space

Qualitative Numeric Planning: Reductions and Complexity