Abstract
Premise selection is a task that selects likely useful premises from a large-scale premise set, which is one of the challenges for automated theorem proving. Nowadays, using graph neural networks to encode logical formulae becomes more and more appealing, as graph representations demonstrate the ability to preserve the semantic and syntactic information of logical formulae. However, graph neural networks in the prior works iteratively update each node’s embedding via information aggregated solely from its neighbors and always ignore the types of nodes. Besides, evidence shows that sharing and exchanging some information between formula pairs can ensure neural-based models more robust in measuring graph relevance. Unluckily, previous graph neural networks generate final graph embeddings independently in the premise selection task. To overcome these shortages, we propose a novel graph neural network, called ARCG-NN, for embedding the first-order logical formulae. The embedding model firstly takes the node types into full consideration and utilizes an attention mechanism based on newly proposed node types to compute weights in the message aggregation. Except for the message from the local neighborhood, ARCG-NN also dynamically exchange the cross-graph information of particular nodes along with the propagation rounds. Besides, an extra gate function on node types is used to filter out irrelevant information in the graph aggregation phase. To train, validate, and test our approach, we build balanced and unbalanced datasets based on the MPTP2078 benchmark. The experimental results demonstrate that the proposed graph neural architecture achieves state-of-the-art classification accuracy on the first-order premise selection task and helps the automated theorem prover E prove more problems in the test dataset.
Similar content being viewed by others
Notes
The command is ./eprover –satauto-schedule –free-numbers -s -R –delete-bad-limit=2000000000 –definitional-cnf=24 –print-statistics –print-version –proof-object –cpu-limit=60 –sine problem_file.
The command is ./eprover –satauto-schedule –free-numbers -s -R –delete-bad-limit=2000000000 –definitional-cnf=24 –print-statistics –print-version –proof-object –cpu-limit=10 problem_file.
References
Alama J, Heskes T, Kühlwein D, Tsivtsivadze E, Urban J (2014) Premise selection for mathematics by corpus analysis and kernel methods. J Autom Reason 52(2):191–213
Chvalovskỳ K (2018) Top-down neural model for formulae. In: International conference on learning representations
Crouse M, Abdelaziz I, Cornelio C, Thost V, Wu L, Forbus K, Fokoue A (2019) Improving graph neural network representations of logical formulae with subgraph pooling. arXiv preprint arXiv:1911.06904
Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering. arXiv preprint arXiv:1606.09375
Evans R, Saxton D, Amos D, Kohli P, Grefenstette E (2018) Can neural networks understand logical entailment? arXiv preprint arXiv:1802.08535
Faizollahzadeh Ardabili S, Najafi B, Shamshirband S, Minaei Bidgoli B, Deo RC, Chau Kw (2018) Computational intelligence approach for modeling hydrogen production: a review. Eng Appl Comput Fluid Mech 12(1):438–458
Fey M, Lenssen JE (2019) Fast graph representation learning with PyTorch geometric. In: ICLR workshop on representation learning on graphs and manifolds
Glimer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE (2017) Neural message passing for quantum chemistry. arXiv preprint arXiv:1704.01212
Hamilton WL, Ying R, Leskovec J (2017) Inductive representation learning on large graphs. arXiv preprint arXiv:1706.02216
Hoder K, Voronkov A (2011) Sine qua non for large theory reasoning. In: International conference on automated deduction, Springer, pp 299–314
Huang D, Dhariwal P, Song D, Sutskever I (2018) Gamepad: a learning environment for theorem proving. arXiv preprint arXiv:1806.00608
Irving G, Szegedy C, Alemi AA, Eén N, Chollet F, Urban J (2016) Deepmath-deep sequence models for premise selection. In: Advances in neural information processing systems, pp 2235–2243
Jakubův J, Urban J (2017) ENIGMA: efficient learning-based inference guiding machine. In: International conference on intelligent computer mathematics, Springer, pp 292–302
Kaliszyk C, Urban J (2015) Mizar 40 for mizar 40. J Autom Reason 55(3):245–256
Kaliszyk C, Urban J, Vyskocil J (2015) Efficient semantic features for automated reasoning over large theories. In: Twenty-fourth international joint conference on artificial intelligence
Kingma D, Ba J (2015) Adam: a method for stochastic optimization. In: International conference on learning representations
Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: International conference on learning representations
Kovács L, Voronkov A (2013) First-order theorem proving and vampire. In: International conference on computer aided verification, Springer, pp 1–35
Kucik AS, Korovin K (2018) Premise selection with neural networks and distributed representation of features. arXiv preprint arXiv:1807.10268
Li Y, Gu C, Dullien T, Vinyals O, Kohli P (2019) Graph matching networks for learning the similarity of graph structured objects. arXiv:1904.12787
Li Y, Tarlow D, Brockschmidt M, Zemel R (2015) Gated graph sequence neural networks. arXiv preprint arXiv:1511.05493
Loos S, Irving G, Szegedy C, Kaliszyk C (2017) Deep network guided proof search. arXiv preprint arXiv:1701.06972
Meng J, Paulson LC (2009) Lightweight relevance filtering for machine-generated resolution problems. J Appl Log 7(1):41–57
Olšák M, Kaliszyk C, Urban J (2019) Property invariant embedding for automated reasoning. arXiv preprint arXiv:1911.12073
Paliwal A, Loos S, Rabe M, Bansal K, Szegedy C (2019) Graph representations for higher-order logic and theorem proving. arXiv preprint arXiv:1905.10006
Paszke A, Gross S, Chintala S, Chanan G, Yang E, DeVito Z, Lin Z, Desmaison A, Antiga L, Lerer A (2017) Automatic differentiation in pytorch. In: NIPS workshop
Piotrowski B, Urban J (2018) ATPboost: learning premise selection in binary setting with ATP feedback. In: International joint conference on automated reasoning, Springer, pp 566–574
Robinson AJ, Voronkov A (2001) Handbook of automated reasoning, vol 1. Gulf Professional Publishing, Oxford
Schulz S (2012) E - a brainiac theorem prover. Ai Commun 15(2):111–126
Shamshirband S, Rabczuk T, Chau KW (2019) A survey of deep learning techniques: application in wind and solar energy resources. IEEE Access 7:164650–164666
Sutcliffe G (2017) The TPTP problem library and associated infrastructure: from CNF to TH0, TPTP v6. 4.0. J Autom Reason 1–20
Taormina R, Chau KW (2015) Ann-based interval forecasting of streamflow discharges using the lube method and mofips. Eng Appl Artif Intell 45:429–440
Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y (2017) Graph attention networks. arXiv preprint arXiv:1710.10903
Wang M, Tang Y, Wang J, Deng J (2017) Premise selection for theorem proving by deep graph embedding. In: Advances in neural information processing systems, pp 2786–2796
Wu C, Chau KW (2013) Prediction of rainfall time series using modular soft computing methods. Eng Appl Artif Intell 26(3):997–1007
Wu F, Souza A, Zhang T, Fifty C, Yu T, Weinberger K (2019) Simplifying graph convolutional networks. In: International conference on machine learning, PMLR, pp 6861–6871
Wu Z, Pan S, Chen F, Long G, Zhang C, Philip SY (2020) A comprehensive survey on graph neural networks. IEEE Trans Neural Netw Learn Syst 32(1):4–24
Yang K, Deng J (2019) Learning to prove theorems via interacting with proof assistants. arXiv preprint arXiv:1905.09381
Zhao L, Akoglu L (2019) Pairnorm: tackling oversmoothing in gnns. arXiv preprint arXiv:1909.12223
Acknowledgements
This research is supported by the National Natural Science Foundation of China (Grant Nos. 61603307, 61673320 and 61473239) and the Grant from MOE (Ministry of Education in China) Project of Humanities and Social Sciences (Grant No. 19YJCZH048).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, Q., Xu, Y. & He, X. Attention recurrent cross-graph neural network for selecting premises. Int. J. Mach. Learn. & Cyber. 13, 1301–1315 (2022). https://doi.org/10.1007/s13042-021-01448-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-021-01448-9