Cargando…
Extracting automata from neural networks using active learning
Deep learning is one of the most advanced forms of machine learning. Most modern deep learning models are based on an artificial neural network, and benchmarking studies reveal that neural networks have produced results comparable to and in some cases superior to human experts. However, the generate...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
PeerJ Inc.
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8064235/ https://www.ncbi.nlm.nih.gov/pubmed/33977128 http://dx.doi.org/10.7717/peerj-cs.436 |
_version_ | 1783682092928860160 |
---|---|
author | Xu, Zhiwu Wen, Cheng Qin, Shengchao He, Mengda |
author_facet | Xu, Zhiwu Wen, Cheng Qin, Shengchao He, Mengda |
author_sort | Xu, Zhiwu |
collection | PubMed |
description | Deep learning is one of the most advanced forms of machine learning. Most modern deep learning models are based on an artificial neural network, and benchmarking studies reveal that neural networks have produced results comparable to and in some cases superior to human experts. However, the generated neural networks are typically regarded as incomprehensible black-box models, which not only limits their applications, but also hinders testing and verifying. In this paper, we present an active learning framework to extract automata from neural network classifiers, which can help users to understand the classifiers. In more detail, we use Angluin’s L* algorithm as a learner and the neural network under learning as an oracle, employing abstraction interpretation of the neural network for answering membership and equivalence queries. Our abstraction consists of value, symbol and word abstractions. The factors that may affect the abstraction are also discussed in the paper. We have implemented our approach in a prototype. To evaluate it, we have performed the prototype on a MNIST classifier and have identified that the abstraction with interval number 2 and block size 1 × 28 offers the best performance in terms of F1 score. We also have compared our extracted DFA against the DFAs learned via the passive learning algorithms provided in LearnLib and the experimental results show that our DFA gives a better performance on the MNIST dataset. |
format | Online Article Text |
id | pubmed-8064235 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | PeerJ Inc. |
record_format | MEDLINE/PubMed |
spelling | pubmed-80642352021-05-10 Extracting automata from neural networks using active learning Xu, Zhiwu Wen, Cheng Qin, Shengchao He, Mengda PeerJ Comput Sci Artificial Intelligence Deep learning is one of the most advanced forms of machine learning. Most modern deep learning models are based on an artificial neural network, and benchmarking studies reveal that neural networks have produced results comparable to and in some cases superior to human experts. However, the generated neural networks are typically regarded as incomprehensible black-box models, which not only limits their applications, but also hinders testing and verifying. In this paper, we present an active learning framework to extract automata from neural network classifiers, which can help users to understand the classifiers. In more detail, we use Angluin’s L* algorithm as a learner and the neural network under learning as an oracle, employing abstraction interpretation of the neural network for answering membership and equivalence queries. Our abstraction consists of value, symbol and word abstractions. The factors that may affect the abstraction are also discussed in the paper. We have implemented our approach in a prototype. To evaluate it, we have performed the prototype on a MNIST classifier and have identified that the abstraction with interval number 2 and block size 1 × 28 offers the best performance in terms of F1 score. We also have compared our extracted DFA against the DFAs learned via the passive learning algorithms provided in LearnLib and the experimental results show that our DFA gives a better performance on the MNIST dataset. PeerJ Inc. 2021-04-19 /pmc/articles/PMC8064235/ /pubmed/33977128 http://dx.doi.org/10.7717/peerj-cs.436 Text en © 2021 Xu et al. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited. |
spellingShingle | Artificial Intelligence Xu, Zhiwu Wen, Cheng Qin, Shengchao He, Mengda Extracting automata from neural networks using active learning |
title | Extracting automata from neural networks using active learning |
title_full | Extracting automata from neural networks using active learning |
title_fullStr | Extracting automata from neural networks using active learning |
title_full_unstemmed | Extracting automata from neural networks using active learning |
title_short | Extracting automata from neural networks using active learning |
title_sort | extracting automata from neural networks using active learning |
topic | Artificial Intelligence |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8064235/ https://www.ncbi.nlm.nih.gov/pubmed/33977128 http://dx.doi.org/10.7717/peerj-cs.436 |
work_keys_str_mv | AT xuzhiwu extractingautomatafromneuralnetworksusingactivelearning AT wencheng extractingautomatafromneuralnetworksusingactivelearning AT qinshengchao extractingautomatafromneuralnetworksusingactivelearning AT hemengda extractingautomatafromneuralnetworksusingactivelearning |