arXiv subjects

Matthew England

Publications and source records attributed to Matthew England.

At least 37 records · Page 2Linked to original sources

arXiv2023

SMT-Solving Induction Proofs of Inequalities

This paper accompanies a new dataset of non-linear real arithmetic problems for the SMT-LIB benchmark collection. The problems come from an automated proof procedure of Gerhold--Kauers, which is well suited for solution by SMT. The problems of this type have not been tackled by SMT-solvers before. We describe the proof technique and give one new such proof to illustrate it. We then describe the dataset and the results of benchmarking. The benchmarks on the new dataset are quite different to the existing ones. The benchmarking also brings forward some interesting debate on the use/inclusion of rational functions and algebraic numbers in the SMT-LIB.

cs.SC↗

arXiv2023

Levelwise construction of a single cylindrical algebraic cell

Satisfiability Modulo Theories (SMT) solvers check the satisfiability of quantifier-free first-order logic formulas. We consider the theory of non-linear real arithmetic where the formulae are logical combinations of polynomial constraints. Here a commonly used tool is the Cylindrical Algebraic Decomposition (CAD) to decompose real space into cells where the constraints are truth-invariant through the use of projection polynomials. An improved approach is to repackage the CAD theory into a search-based algorithm: one that guesses sample points to satisfy the formula, and generalizes guesses that conflict constraints to cylindrical cells around samples which are avoided in the continuing search. Such an approach can lead to a satisfying assignment more quickly, or conclude unsatisfiability with fewer cells. A notable example of this approach is Jovanović and de Moura's NLSAT algorithm. Since these cells are produced locally to a sample we might need fewer projection polynomials than the traditional CAD projection. The original NLSAT algorithm reduced the set a little; while Brown's single cell construction reduced it much further still. However, the shape and size of the cell produced depends on the order in which the polynomials are considered. This paper proposes a method to construct such cells levelwise, i.e. built level-by-level according to a variable ordering. We still use a reduced number of projection polynomials, but can now consider a variety of different reductions and use heuristics to select the projection polynomials in order to optimise the shape of the cell under construction. We formulate all the necessary theory as a proof system: while not a common presentation for work in this field, it allows an elegant decoupling of heuristics from the algorithm and its proof of correctness.

cs.SC↗

arXiv2023

Data Augmentation for Mathematical Objects

This paper discusses and evaluates ideas of data balancing and data augmentation in the context of mathematical objects: an important topic for both the symbolic computation and satisfiability checking communities, when they are making use of machine learning techniques to optimise their tools. We consider a dataset of non-linear polynomial problems and the problem of selecting a variable ordering for cylindrical algebraic decomposition to tackle these with. By swapping the variable names in already labelled problems, we generate new problem instances that do not require any further labelling when viewing the selection as a classification problem. We find this augmentation increases the accuracy of ML models by 63% on average. We study what part of this improvement is due to the balancing of the dataset and what is achieved thanks to further increasing the size of the dataset, concluding that both have a very significant effect. We finish the paper by reflecting on how this idea could be applied in other uses of machine learning in mathematics.

cs.SC↗

arXiv2023

Generating Elementary Integrable Expressions

There has been an increasing number of applications of machine learning to the field of Computer Algebra in recent years, including to the prominent sub-field of Symbolic Integration. However, machine learning models require an abundance of data for them to be successful and there exist few benchmarks on the scale required. While methods to generate new data already exist, they are flawed in several ways which may lead to bias in machine learning models trained upon them. In this paper, we describe how to use the Risch Algorithm for symbolic integration to create a dataset of elementary integrable expressions. Further, we show that data generated this way alleviates some of the flaws found in earlier methods.

cs.SC↗

arXiv2022

Polynomial Superlevel Set Representation of the Multistationarity Region of Chemical Reaction Networks

In this paper we introduce a new representation for the multistationarity region of a reaction network, using polynomial superlevel sets. The advantages of using this polynomial superlevel set representation over the already existing representations (cylindrical algebraic decompositions, numeric sampling, rectangular divisions) is discussed, and algorithms to compute this new representation are provided. The results are given for the general mathematical formalism of a parametric system of equations and so may be applied to other application domains.

q-bio.MN↗

arXiv2022

SC-Square: Overview to 2021

This extended abstract was written to accompany an invited talk at the 2021 SC-Square Workshop, where the author was asked to give an overview of SC-Square progress to date. The author first reminds the reader of the definition of SC-Square, then briefly outlines some of the history, before picking out some (personal) scientific highlights.

cs.SC↗

arXiv2022

SC-Square: Future Progress with Machine Learning?

The algorithms employed by our communities are often underspecified, and thus have multiple implementation choices, which do not effect the correctness of the output, but do impact the efficiency or even tractability of its production. In this extended abstract, to accompany a keynote talk at the 2021 SC-Square Workshop, we survey recent work (both the author's and from the literature) on the use of Machine Learning technology to improve algorithms of interest to SC-Square.

cs.SC↗

arXiv2022

New heuristic to choose a cylindrical algebraic decomposition variable ordering motivated by complexity analysis

It is well known that the variable ordering can be critical to the efficiency or even tractability of the cylindrical algebraic decomposition (CAD) algorithm. We propose new heuristics inspired by complexity analysis of CAD to choose the variable ordering. These heuristics are evaluated against existing heuristics with experiments on the SMT-LIB benchmarks using both existing performance metrics and a new metric we propose for the problem at hand. The best of these new heuristics chooses orderings that lead to timings on average 17% slower than the virtual-best: an improvement compared to the prior state-of-the-art which achieved timings 25% slower.

cs.SC↗

arXiv2022

Feedback and Engagement on an Introductory Programming Module

We ran a study on engagement and achievement for a first year undergraduate programming module which used an online learning environment containing tasks which generate automated feedback. Students could also access human feedback from traditional labs. We gathered quantitative data on engagement and achievement which allowed us to split the cohort into 6 groups. We then ran interviews with students after the end of the module to produce qualitative data on perceptions of what feedback is, how useful it is, the uses made of it, and how it bears on engagement. A general finding was that human and automated feedback are different but complementary. However there are different feedback needs by group. Our findings imply: (1) that a blended human-automated feedback approach improves engagement; and (2) that this approach needs to be differentiated according to type of student. We give implications for the design of feedback for programming modules.

cs.CY↗

arXiv2021

Proving UNSAT in SMT: The Case of Quantifier Free Non-Linear Real Arithmetic

We discuss the topic of unsatisfiability proofs in SMT, particularly with reference to quantifier free non-linear real arithmetic. We outline how the methods here do not admit trivial proofs and how past formalisation attempts are not sufficient. We note that the new breed of local search based algorithms for this domain may offer an easier path forward.

cs.LO↗

arXiv2020

Deciding the Consistency of Non-Linear Real Arithmetic Constraints with a Conflict Driven Search Using Cylindrical Algebraic Coverings

We present a new algorithm for determining the satisfiability of conjunctions of non-linear polynomial constraints over the reals, which can be used as a theory solver for satisfiability modulo theory (SMT) solving for non-linear real arithmetic. The algorithm is a variant of Cylindrical Algebraic Decomposition (CAD) adapted for satisfiability, where solution candidates (sample points) are constructed incrementally, either until a satisfying sample is found or sufficient samples have been sampled to conclude unsatisfiability. The choice of samples is guided by the input constraints and previous conflicts. The key idea behind our new approach is to start with a partial sample; demonstrate that it cannot be extended to a full sample; and from the reasons for that rule out a larger space around the partial sample, which build up incrementally into a cylindrical algebraic covering of the space. There are similarities with the incremental variant of CAD, the NLSAT method of Jovanovic and de Moura, and the NuCAD algorithm of Brown; but we present worked examples and experimental results on a preliminary implementation to demonstrate the differences to these, and the benefits of the new approach.

cs.SC↗

arXiv2020

A machine learning based software pipeline to pick the variable ordering for algorithms with polynomial inputs

We are interested in the application of Machine Learning (ML) technology to improve mathematical software. It may seem that the probabilistic nature of ML tools would invalidate the exact results prized by such software, however, the algorithms which underpin the software often come with a range of choices which are good candidates for ML application. We refer to choices which have no effect on the mathematical correctness of the software, but do impact its performance. In the past we experimented with one such choice: the variable ordering to use when building a Cylindrical Algebraic Decomposition (CAD). We used the Python library Scikit-Learn (sklearn) to experiment with different ML models, and developed new techniques for feature generation and hyper-parameter selection. These techniques could easily be adapted for making decisions other than our immediate application of CAD variable ordering. Hence in this paper we present a software pipeline to use sklearn to pick the variable ordering for an algorithm that acts on a polynomial system. The code described is freely available online.

cs.SC↗

arXiv2020

New Opportunities for the Formal Proof of Computational Real Geometry?

The purpose of this paper is to explore the question "to what extent could we produce formal, machine-verifiable, proofs in real algebraic geometry?" The question has been asked before but as yet the leading algorithms for answering such questions have not been formalised. We present a thesis that a new algorithm for ascertaining satisfiability of formulae over the reals via Cylindrical Algebraic Coverings [Ábrahám, Davenport, England, Kremer, \emph{Deciding the Consistency of Non-Linear Real Arithmetic Constraints with a Conflict Driver Search Using Cylindrical Algebraic Coverings}, 2020] might provide trace and outputs that allow the results to be more susceptible to machine verification than those of competing algorithms.

cs.SC↗

arXiv2019

Computing with CodeRunner at Coventry University: Automated summative assessment of Python and C++ code

CodeRunner is a free open-source Moodle plugin for automatically marking student code. We describe our experience using CodeRunner for summative assessment in our first year undergraduate programming curriculum at Coventry University. We use it to assess both Python3 and C++14 code (CodeRunner supports other languages also). We give examples of our questions and report on how key metrics have changed following its use at Coventry.

cs.CY↗

arXiv2019

Improved cross-validation for classifiers that make algorithmic choices to minimise runtime without compromising output correctness

Our topic is the use of machine learning to improve software by making choices which do not compromise the correctness of the output, but do affect the time taken to produce such output. We are particularly concerned with computer algebra systems (CASs), and in particular, our experiments are for selecting the variable ordering to use when performing a cylindrical algebraic decomposition of $n$-dimensional real space with respect to the signs of a set of polynomials. In our prior work we explored the different ML models that could be used, and how to identify suitable features of the input polynomials. In the present paper we both repeat our prior experiments on problems which have more variables (and thus exponentially more possible orderings), and examine the metric which our ML classifiers targets. The natural metric is computational runtime, with classifiers trained to pick the ordering which minimises this. However, this leads to the situation were models do not distinguish between any of the non-optimal orderings, whose runtimes may still vary dramatically. In this paper we investigate a modification to the cross-validation algorithms of the classifiers so that they do distinguish these cases, leading to improved results.

cs.SC↗

arXiv2019

First Year Computer Science Projects at Coventry University: Activity-led integrative team projects with continuous assessment

We describe the group projects undertaken by first year undergraduate Computer Science students at Coventry University. These are integrative course projects: designed to bring together the topics from the various modules students take, to apply them as a coherent whole. They follow an activity-led approach, with students given a loose brief and a lot of freedom in how to develop their project. We outline the new regulations at Coventry University which eases the use of such integrative projects. We then describe our continuous assessment approach: where students earn a weekly mark by demonstrating progress to a teacher as an open presentation to the class. It involves a degree of self and peer assessment and allows for an assessment of group work that is both fair, and seen to be fair. It builds attendance, self-study / continuous engagement habits, public speaking / presentation skills, and rewards group members for making meaningful individual contributions.

cs.CY↗

arXiv2019

Comparing machine learning models to choose the variable ordering for cylindrical algebraic decomposition

There has been recent interest in the use of machine learning (ML) approaches within mathematical software to make choices that impact on the computing performance without affecting the mathematical correctness of the result. We address the problem of selecting the variable ordering for cylindrical algebraic decomposition (CAD), an important algorithm in Symbolic Computation. Prior work to apply ML on this problem implemented a Support Vector Machine (SVM) to select between three existing human-made heuristics, which did better than anyone heuristic alone. The present work extends to have ML select the variable ordering directly, and to try a wider variety of ML techniques. We experimented with the NLSAT dataset and the Regular Chains Library CAD function for Maple 2018. For each problem, the variable ordering leading to the shortest computing time was selected as the target class for ML. Features were generated from the polynomial input and used to train the following ML models: k-nearest neighbours (KNN) classifier, multi-layer perceptron (MLP), decision tree (DT) and SVM, as implemented in the Python scikit-learn package. We also compared these with the two leading human constructed heuristics for the problem: Brown's heuristic and sotd. On this dataset all of the ML approaches outperformed the human made heuristics, some by a large margin.

cs.SC↗

arXiv2019

Algorithmically generating new algebraic features of polynomial systems for machine learning

There are a variety of choices to be made in both computer algebra systems (CASs) and satisfiability modulo theory (SMT) solvers which can impact performance without affecting mathematical correctness. Such choices are candidates for machine learning (ML) approaches, however, there are difficulties in applying standard ML techniques, such as the efficient identification of ML features from input data which is typically a polynomial system. Our focus is selecting the variable ordering for cylindrical algebraic decomposition (CAD), an important algorithm implemented in several CASs, and now also SMT-solvers. We created a framework to describe all the previously identified ML features for the problem and then enumerated all options in this framework to automatically generation many more features. We validate the usefulness of these with an experiment which shows that an ML choice for CAD variable ordering is superior to those made by human created heuristics, and further improved with these additional features. We expect that this technique of feature generation could be useful for other choices related to CAD, or even choices for other algorithms with polynomial systems for input.

cs.SC↗