Planning with tensor networks based on active inference

被引：0

作者：

Wauthier, Samuel T. ^{[1
]}

Verbelen, Tim ^{[2
]}

Dhoedt, Bart ^{[1
]}

Vanhecke, Bram ^{[3
]}

机构：

[1] IDLab, Department of Information Technology, Ghent University—imec, Technologiepark-Zwijnaarde 126, Ghent,B-9052, Belgium

[2] VERSES AI Research Lab, Los Angeles,CA,90016, United States

[3] Faculty of Physics, Faculty of Mathematics, Quantum Optics, Quantum Nanophysics and Quantum Information, University of Vienna, Boltzmanngasse 5, Vienna,1090, Austria

来源：

Machine Learning: Science and Technology | 2024年 / 5卷 / 04期

关键词：

Tensor networks (TNs) have seen an increase in applications in recent years. While they were originally developed to model many-body quantum systems; their usage has expanded into the field of machine learning. This work adds to the growing range of applications by focusing on planning by combining the generative modeling capabilities of matrix product states and the action selection algorithm provided by active inference. Their ability to deal with the curse of dimensionality; to represent probability distributions; and to dynamically discover hidden variables make matrix product states specifically an interesting choice to use as the generative model in active inference; which relies on ‘beliefs’ about hidden states within an environment. We evaluate our method on the T-maze and Frozen Lake environments; and show that the TN-based agent acts Bayes optimally as expected under active inference. © 2024 The Author(s). Published by IOP Publishing Ltd;

D O I：

10.1088/2632-2153/ad7571

中图分类号：

学科分类号：

摘要：

引用

共 50 条

[1] Learning Generative Models for Active Inference Using Tensor Networks
Wauthier, Samuel T.
Vanhecke, Bram
Verbelen, Tim
Dhoedt, Bart
ACTIVE INFERENCE, IWAI 2022, 2023, 1721 : 285 - 297
[2] Planning and navigation as active inference
Kaplan, Raphael
Friston, Karl J.
BIOLOGICAL CYBERNETICS, 2018, 112 (04) : 323 - 343
[3] Planning and navigation as active inference
Raphael Kaplan
Karl J. Friston
Biological Cybernetics, 2018, 112 : 323 - 343
[4] Dynamic planning in hierarchical active inference
Priorelli, Matteo
Stoianov, Ivilin Peev
Neural Networks, 2025, 185
[5] Application of production active planning based on neural networks
Huang, Jingping
Feng, Shan
Zhou, Kaibo
Huazhong Ligong Daxue Xuebao/Journal Huazhong (Central China) University of Science and Technology, 2000, 28 (08): : 27 - 29
[6] Probabilistic inference in the era of tensor networks and differential programming
Roa-Villescas, Martin
Gao, Xuanzhao
Stuijk, Sander
Corporaal, Henk
Liu, Jin-Guo
PHYSICAL REVIEW RESEARCH, 2024, 6 (03):
[7] On Predictive Planning and Counterfactual Learning in Active Inference
Paul, Aswin
Isomura, Takuya
Razi, Adeel
ENTROPY, 2024, 26 (06)
[8] LEARNING PERCEPTION AND PLANNING WITH DEEP ACTIVE INFERENCE
Catal, Ozan
Verbelen, Tim
Nauta, Johannes
De Boom, Cedric
Dhoedt, Bart
2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 3952 - 3956
[9] Sniffer-based Inference of the Causes of Active Scanning in WiFi Networks
Jaisinghani, Dheryta
Naik, Vinayak
Kaul, Sanjit K.
Roy, Sumit
2017 TWENTY-THIRD NATIONAL CONFERENCE ON COMMUNICATIONS (NCC), 2017,
[10] Control Flow in Active Inference Systems-Part II: Tensor Networks as General Models of Control Flow
Fields, Chris
Fabrocini, Filippo
Friston, Karl
Glazebrook, James F.
Hazan, Hananel
Levin, Michael
Marciano, Antonino
IEEE TRANSACTIONS ON MOLECULAR BIOLOGICAL AND MULTI-SCALE COMMUNICATIONS, 2023, 9 (02): : 246 - 256

← 1 2 3 4 5 →