Property Invariant Embedding for Automated Reasoning

@inproceedings{Olsk2020PropertyIE,
  title={Property Invariant Embedding for Automated Reasoning},
  author={Miroslav Ols{\'a}k and C. Kaliszyk and Josef Urban},
  booktitle={ECAI},
  year={2020}
}
Automated reasoning and theorem proving have recently become major challenges for machine learning. In other domains, representations that are able to abstract over unimportant transformations, such as abstraction over translations and rotations in vision, are becoming more common. Standard methods of embedding mathematical formulas for learning theorem proving are however yet unable to handle many important transformations. In particular, embedding previously unseen labels, that often arise in… 
View With Semantic Reader
26 Citations
Highly Influential Citations
3
Background Citations
9
Methods Citations
8

Figures and Tables from this paper

26 Citations

Citation Type

Citation Type

A Study of Continuous Vector Representationsfor Theorem Proving
TLDR
This paper develops an encoding that allows for logical properties to be preserved and is additionally reversible, which means that the tree shape of a formula including all symbols can be reconstructed from the dense vector representation.
An Experimental Study of Formula Embeddings for Automated Theorem Proving in First-Order Logic
TLDR
This paper study and experimentally compare pattern-based embeddings that are applied in current systems with popular graph-based encodings, most of which have not been considered in the theorem proving context before, and presents a detailed analysis across several dimensions of theorem prover performance beyond just proof completion rate.
Improving Graph Neural Network Representations of Logical Formulae with Subgraph Pooling
TLDR
This work proposes a novel approach for embedding logical formulae that is designed to overcome the representational limitations of prior approaches and achieves state-of-the-art performance on both premise selection and proof step classification.
Exploring Representation of Horn Clauses using GNNs
TLDR
This work considers Constrained Horn Clauses (CHCs) as a standard representation of program verification problems, and proposes a new Relational Hypergraph Neural Network (R-HyGNN) architecture, extending Relational Graph Convolutional Networks, to handle hypergraphs.
Adversarial Learning to Reason in an Arbitrary Logic
TLDR
This work proposes Monte-Carlo simulations guided by reinforcement learning that can work in an arbitrarily specified logic, without any human knowledge or set of problems, and practically demonstrates the feasibility of the approach in multiple logical systems.
The Role of Entropy in Guiding a Connection Prover
TLDR
This work starts by incorporating a state-of-the-art learning algorithm – a graph neural network (GNN) – into the plCoP theorem prover, and shows that a proper entropy regularization, i.e., training the GNN not to be overconfident, greatly improves pl coP’s performance on a large mathematical corpus.
ENIGMA Anonymous: Symbol-Independent Inference Guiding Machine (System Description)
TLDR
An implementation of gradient boosting and neural guidance of saturation-style automated theorem provers that does not depend on consistent symbol names across problems is described and is evaluated on the MPTP large-theory benchmark and shown to achieve comparable real-time performance to state-of-the-art symbol-based methods.
A Deep Reinforcement Learning Approach to First-Order Logic Theorem Proving
TLDR
TRAIL is introduced, a system that applies deep reinforcement learning to saturation-based theorem proving and leverages a novel neural representation of the state of a theorem prover and a novel characterization of the inference selection process in terms of an attention-based action policy.
A Deep Reinforcement Learning based Approach to Learning Transferable Proof Guidance Strategies
TLDR
It is shown that TRAIL's learned strategies provide a comparable performance to an established heuristics-based theorem prover, suggesting that the neural architecture in TRAIL is well suited for representing and processing of logical formalisms.
Learning Theorem Proving Components
TLDR
This work describes several algorithms and experiments with ENIGMA, advancing the idea of contextual evaluation based on learning important components of the graph of clauses, and equipping the E/ENIGMA system with a graph neural network that chooses the next given clause based on its evaluation in the context of previously selected clauses.
...
...

References

SHOWING 1-10 OF 31 REFERENCES
Efficient Semantic Features for Automated Reasoning over Large Theories
TLDR
This work proposes novel semantic features characterizing the statements in such large semantic knowledge bases and carries out their efficient implementation using deductive-AI data-structures such as substitution trees and discrimination nets, and shows that they significantly improve the strength of existing knowledge selection methods and automated reasoning methods over the large formal knowledge bases.
Learning search control knowledge for equational deduction
TLDR
This thesis develops techniques to automatically learn good search heuristics to control the proof search of a superposition-based theorem prover for clausal logic with equality and describes a variant of the superposition calculus and an efficient proof procedure implementing this calculus.
HolStep: A Machine Learning Dataset for Higher-order Logic Theorem Proving
TLDR
A new dataset based on Higher-Order Logic (HOL) proofs is introduced, for the purpose of developing new machine learning-based theorem-proving strategies and the results of these models show the promise of applying machine learning to HOL theorem proving.
Premise Selection for Theorem Proving by Deep Graph Embedding
We propose a deep learning-based approach to the problem of premise selection: selecting mathematical statements relevant for proving a given conjecture. We represent a higher-order logic formula as…
Learning Continuous Semantic Representations of Symbolic Expressions
TLDR
An exhaustive evaluation on the task of checking equivalence on a highly diverse class of symbolic algebraic and boolean expression types is performed, showing that the proposed neural equivalence networks model significantly outperforms existing architectures.
Aligning concepts across proof assistant libraries
Deep Network Guided Proof Search
TLDR
Experimental evidence is given that with a hybrid, two-phase approach, deep learning based guidance can significantly reduce the average number of proof search steps while increasing the number of theorems proved.
Reinforcement Learning of Theorem Proving
TLDR
A theorem proving algorithm that uses practically no domain heuristics for guiding its connection-style proof search and solves within the same number of inferences over 40% more problems than a baseline prover, which is an unusually high improvement in this hard AI domain.
Premise selection with neural networks and distributed representation of features
TLDR
This work presents the problem of selecting relevant premises for a proof of a given statement, and uses dimensionality reduction technique, to replace long and sparse signature vectors with their compact and dense embedded versions.
Premise Selection and External Provers for HOL4
TLDR
An add-on to the HOL4 proof assistant and an adaptation of the HOL(y)Hammer system that provides machine learning-based premise selection and automated reasoning also for HOL4, which directly benefits HOL4 users by automatically finding proofs dependencies that can be reconstructed by Metis.
...
...