Computer Science > Artificial Intelligence
[Submitted on 26 Nov 2018 (v1), last revised 6 Mar 2020 (this version, v12)]
Title:Compact and Efficient Encodings for Planning in Factored State and Action Spaces with Learned Binarized Neural Network Transition Models
View PDFAbstract:In this paper, we leverage the efficiency of Binarized Neural Networks (BNNs) to learn complex state transition models of planning domains with discretized factored state and action spaces. In order to directly exploit this transition structure for planning, we present two novel compilations of the learned factored planning problem with BNNs based on reductions to Weighted Partial Maximum Boolean Satisfiability (FD-SAT-Plan+) as well as Binary Linear Programming (FD-BLP-Plan+). Theoretically, we show that our SAT-based Bi-Directional Neuron Activation Encoding is asymptotically the most compact encoding relative to the current literature and supports Unit Propagation (UP) -- an important property that facilitates efficiency in SAT solvers. Experimentally, we validate the computational efficiency of our Bi-Directional Neuron Activation Encoding in comparison to an existing neuron activation encoding and demonstrate the ability to learn complex transition models with BNNs. We test the runtime efficiency of both FD-SAT-Plan+ and FD-BLP-Plan+ on the learned factored planning problem showing that FD-SAT-Plan+ scales better with increasing BNN size and complexity. Finally, we present a finite-time incremental constraint generation algorithm based on generalized landmark constraints to improve the planning accuracy of our encodings through simulated or real-world interaction.
Submission history
From: Buser Say [view email][v1] Mon, 26 Nov 2018 14:59:29 UTC (1,208 KB)
[v2] Tue, 27 Nov 2018 18:48:51 UTC (1,208 KB)
[v3] Wed, 28 Nov 2018 02:08:22 UTC (1,208 KB)
[v4] Thu, 29 Nov 2018 15:31:00 UTC (1,209 KB)
[v5] Fri, 30 Nov 2018 17:15:31 UTC (1,209 KB)
[v6] Thu, 6 Dec 2018 18:13:01 UTC (1,210 KB)
[v7] Fri, 7 Dec 2018 14:38:28 UTC (1,210 KB)
[v8] Mon, 10 Dec 2018 02:21:17 UTC (1,210 KB)
[v9] Thu, 10 Jan 2019 11:25:20 UTC (1,210 KB)
[v10] Tue, 9 Apr 2019 00:23:16 UTC (1,210 KB)
[v11] Thu, 3 Oct 2019 13:08:37 UTC (1,518 KB)
[v12] Fri, 6 Mar 2020 17:47:58 UTC (2,587 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.