dtControl: decision tree learning algorithms for controller
representation

D 2020

dtControl: decision tree learning algorithms for controller representation

ASHOK, Pranav; Mathias JACKERMEIER; Pushpak JAGTAP; Jan KŘETÍNSKÝ; Maximilian WEININGER et al.

Základní údaje

Originální název

dtControl: decision tree learning algorithms for controller representation

Autoři

ASHOK, Pranav; Mathias JACKERMEIER; Pushpak JAGTAP; Jan KŘETÍNSKÝ; Maximilian WEININGER a Majid ZAMANI

Vydání

HSCC '20: 23rd ACM International Conference on Hybrid Systems: Computation and Control, Sydney, New South Wales, Australia, April 21-24, 2020, od s. 1-7, 7 s. 2020

Nakladatel

ACM

Další údaje

Typ výsledku

Stať ve sborníku

Označené pro přenos do RIV

Organizační jednotka

Fakulta informatiky

ISBN

9781450370189

DOI

https://doi.org/10.1145/3365365.3382220

Změněno: 17. 3. 2025 14:43, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Decision tree learning is a popular classification technique most commonly used in machine learning applications. Recent work has shown that decision trees can be used to represent provably-correct controllers concisely. Compared to representations using lookup tables or binary decision diagrams, decision trees are smaller and more explainable. We present dtControl, an easily extensible tool for representing memoryless controllers as decision trees. We give a comprehensive evaluation of various decision tree learning algorithms applied to 10 case studies arising out of correct-by-construction controller synthesis. These algorithms include two new techniques, one for using arbitrary linear binary classifiers in the decision tree learning, and one novel approach for determinizing controllers during the decision tree construction. In particular the latter turns out to be extremely efficient, yielding decision trees with a single-digit number of decision nodes on 5 of the case studies.

Citovat

ASHOK, Pranav; Mathias JACKERMEIER; Pushpak JAGTAP; Jan KŘETÍNSKÝ; Maximilian WEININGER a Majid ZAMANI. dtControl: decision tree learning algorithms for controller representation. In HSCC '20: 23rd ACM International Conference on Hybrid Systems: Computation and Control, Sydney, New South Wales, Australia, April 21-24, 2020. ACM, 2020, s. 1-7. ISBN 9781450370189. Dostupné z: https://doi.org/10.1145/3365365.3382220.

@inproceedings{2484792,
   author = {Ashok, Pranav and Jackermeier, Mathias and Jagtap, Pushpak and Křetínský, Jan and Weininger, Maximilian and Zamani, Majid},
   booktitle = {HSCC '20: 23rd ACM International Conference on Hybrid Systems: Computation and Control, Sydney, New South Wales, Australia, April 21-24, 2020},
   doi = {https://doi.org/10.1145/3365365.3382220},
   isbn = {9781450370189},
   pages = {1-7},
   publisher = {ACM},
   title = {dtControl: decision tree learning algorithms for controller representation},
   year = {2020}
}

TY  - CONF
ID  - 2484792
AU  - Ashok, Pranav - Jackermeier, Mathias - Jagtap, Pushpak - Křetínský, Jan - Weininger, Maximilian - Zamani, Majid
PY  - 2020
TI  - dtControl: decision tree learning algorithms for controller representation
PB  - ACM
SN  - 9781450370189
N2  - Decision tree learning is a popular classification technique most commonly used in machine learning applications. Recent work has shown that decision trees can be used to represent provably-correct controllers concisely. Compared to representations using lookup tables or binary decision diagrams, decision trees are smaller and more explainable. We present dtControl, an easily extensible tool for representing memoryless controllers as decision trees. We give a comprehensive evaluation of various decision tree learning algorithms applied to 10 case studies arising out of correct-by-construction controller synthesis. These algorithms include two new techniques, one for using arbitrary linear binary classifiers in the decision tree learning, and one novel approach for determinizing controllers during the decision tree construction. In particular the latter turns out to be extremely efficient, yielding decision trees with a single-digit number of decision nodes on 5 of the case studies.
ER  -

ASHOK, Pranav; Mathias JACKERMEIER; Pushpak JAGTAP; Jan KŘETÍNSKÝ; Maximilian WEININGER a Majid ZAMANI. dtControl: decision tree learning algorithms for controller representation. In \textit{HSCC '20: 23rd ACM International Conference on Hybrid Systems: Computation and Control, Sydney, New South Wales, Australia, April 21-24, 2020}. ACM, 2020, s.~1-7. ISBN~9781450370189. Dostupné z: https://doi.org/10.1145/3365365.3382220.

Přehled o publikaci