Extracting automata from recurrent neural networks using queries and counterexamples

Gail Weiss; Yoav Goldberg; Eran Yahav

Extracting automata from recurrent neural networks using queries and counterexamples

Gail Weiss
, Yoav Goldberg
, Eran Yahav

Technion-Israel Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

46 Scopus citations

Abstract

We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L^∗ algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.

Original language	English
Title of host publication	35th International Conference on Machine Learning, ICML 2018
Editors	Jennifer Dy, Andreas Krause
Publisher	International Machine Learning Society (IMLS)
Pages	8340-8349
Number of pages	10
ISBN (Electronic)	9781510867963
State	Published - 2018
Externally published	Yes
Event	35th International Conference on Machine Learning, ICML 2018 - Stockholm, Sweden Duration: 10 Jul 2018 → 15 Jul 2018

Publication series

Name	35th International Conference on Machine Learning, ICML 2018
Volume	12

Conference

Conference	35th International Conference on Machine Learning, ICML 2018
Country/Territory	Sweden
City	Stockholm
Period	10/07/18 → 15/07/18

Bibliographical note

Publisher Copyright:
© 35th International Conference on Machine Learning, ICML 2018.All Rights Reserved.

Cite this

@inproceedings{17e6d6e705504b9ca29363842a20a934,

title = "Extracting automata from recurrent neural networks using queries and counterexamples",

abstract = "We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L∗ algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.",

author = "Gail Weiss and Yoav Goldberg and Eran Yahav",

note = "Publisher Copyright: {\textcopyright} 35th International Conference on Machine Learning, ICML 2018.All Rights Reserved.; 35th International Conference on Machine Learning, ICML 2018 ; Conference date: 10-07-2018 Through 15-07-2018",

year = "2018",

language = "אנגלית",

series = "35th International Conference on Machine Learning, ICML 2018",

publisher = "International Machine Learning Society (IMLS)",

pages = "8340--8349",

editor = "Jennifer Dy and Andreas Krause",

booktitle = "35th International Conference on Machine Learning, ICML 2018",

}

Weiss, G, Goldberg, Y & Yahav, E 2018, Extracting automata from recurrent neural networks using queries and counterexamples. in J Dy & A Krause (eds), 35th International Conference on Machine Learning, ICML 2018. 35th International Conference on Machine Learning, ICML 2018, vol. 12, International Machine Learning Society (IMLS), pp. 8340-8349, 35th International Conference on Machine Learning, ICML 2018, Stockholm, Sweden, 10/07/18.

Extracting automata from recurrent neural networks using queries and counterexamples. / Weiss, Gail; Goldberg, Yoav; Yahav, Eran.
35th International Conference on Machine Learning, ICML 2018. ed. / Jennifer Dy; Andreas Krause. International Machine Learning Society (IMLS), 2018. p. 8340-8349 (35th International Conference on Machine Learning, ICML 2018; Vol. 12).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Extracting automata from recurrent neural networks using queries and counterexamples

AU - Weiss, Gail

AU - Goldberg, Yoav

AU - Yahav, Eran

PY - 2018

Y1 - 2018

N2 - We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L∗ algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.

AB - We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L∗ algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.

UR - https://www.scopus.com/pages/publications/85057317818

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85057317818

T3 - 35th International Conference on Machine Learning, ICML 2018

SP - 8340

EP - 8349

BT - 35th International Conference on Machine Learning, ICML 2018

A2 - Dy, Jennifer

A2 - Krause, Andreas

PB - International Machine Learning Society (IMLS)

T2 - 35th International Conference on Machine Learning, ICML 2018

Y2 - 10 July 2018 through 15 July 2018

ER -

Extracting automata from recurrent neural networks using queries and counterexamples

Abstract

Publication series

Conference

Bibliographical note

Other files and links

Fingerprint

Cite this