Deep Reinforcement Learning for Synthesizing Functions in Higher-Order Logic

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F20%3A00346127" target="_blank" >RIV/68407700:21730/20:00346127 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.29007/7jmg" target="_blank" >https://doi.org/10.29007/7jmg</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.29007/7jmg" target="_blank" >10.29007/7jmg</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Deep Reinforcement Learning for Synthesizing Functions in Higher-Order Logic

Popis výsledku v původním jazyce

The paper describes a deep reinforcement learning framework based on self-supervised learning within the proof assistant HOL4. A close interaction between the machine learning modules and the HOL4 library is achieved by the choice of tree neural networks (TNNs) as machine learning models and the internal use of HOL4 terms to represent tree structures of TNNs. Recursive improvement is possible when a task is expressed as a search problem. In this case, a Monte Carlo Tree Search (MCTS) algorithm guided by a TNN can be used to explore the search space and produce better examples for training the next TNN. As an illustration, term synthesis tasks on combinators and Diophantine equations are specified and learned. We achieve a success rate of 65% on combinator synthesis problems outperforming state-of-the-art ATPs run with their best general set of strategies. We set a precedent for statistically guided synthesis of Diophantine equations by solving 78.5% of the generated test problems

Název v anglickém jazyce

Deep Reinforcement Learning for Synthesizing Functions in Higher-Order Logic

Popis výsledku anglicky

The paper describes a deep reinforcement learning framework based on self-supervised learning within the proof assistant HOL4. A close interaction between the machine learning modules and the HOL4 library is achieved by the choice of tree neural networks (TNNs) as machine learning models and the internal use of HOL4 terms to represent tree structures of TNNs. Recursive improvement is possible when a task is expressed as a search problem. In this case, a Monte Carlo Tree Search (MCTS) algorithm guided by a TNN can be used to explore the search space and produce better examples for training the next TNN. As an illustration, term synthesis tasks on combinators and Diophantine equations are specified and learned. We achieve a success rate of 65% on combinator synthesis problems outperforming state-of-the-art ATPs run with their best general set of strategies. We set a precedent for statistically guided synthesis of Diophantine equations by solving 78.5% of the generated test problems

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

R - Projekt Ramcoveho programu EK

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

EPiC Series in Computing

ISBN

—

ISSN

2398-7340

e-ISSN

2398-7340

Počet stran výsledku

19

Strana od-do

230-248

Název nakladatele

EasyChair Publications

Místo vydání

Manchester

Místo konání akce

Alicante

Datum konání akce

22. 5. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—