Searcharxiv⌕ Search

arXiv subjects

Ekaterina Komendantskaya

Publications and source records attributed to Ekaterina Komendantskaya.

At least 37 records · Page 2Linked to original sources
arXiv2022

CheckINN: Wide Range Neural Network Verification in Imandra (Extended)

Neural networks are increasingly relied upon as components of complex safety-critical systems such as autonomous vehicles. There is high demand for tools and methods that embed neural network verification in a larger verification cycle. However, neural network verification is difficult due to a wide range of verification properties of interest, each typically only amenable to verification in specialised solvers. In this paper, we show how Imandra, a functional programming language and a theorem prover originally designed for verification, validation and simulation of financial infrastructure can offer a holistic infrastructure for neural network verification. We develop a novel library CheckINN that formalises neural networks in Imandra, and covers different important facets of neural network verification.

cs.LO↗
arXiv2022

Neural Networks in Imandra: Matrix Representation as a Verification Choice

The demand for formal verification tools for neural networks has increased as neural networks have been deployed in a growing number of safety-critical applications. Matrices are a data structure essential to formalising neural networks. Functional programming languages encourage diverse approaches to matrix definitions. This feature has already been successfully exploited in different applications. The question we ask is whether, and how, these ideas can be applied in neural network verification. A functional programming language Imandra combines the syntax of a functional programming language and the power of an automated theorem prover. Using these two key features of Imandra, we explore how different implementations of matrices can influence automation of neural network verification.

cs.LO↗
arXiv2022

Differentiable Logics for Neural Network Training and Verification

The rising popularity of neural networks (NNs) in recent years and their increasing prevalence in real-world applications have drawn attention to the importance of their verification. While verification is known to be computationally difficult theoretically, many techniques have been proposed for solving it in practice. It has been observed in the literature that by default neural networks rarely satisfy logical constraints that we want to verify. A good course of action is to train the given NN to satisfy said constraint prior to verifying them. This idea is sometimes referred to as continuous verification, referring to the loop between training and verification. Usually training with constraints is implemented by specifying a translation for a given formal logic language into loss functions. These loss functions are then used to train neural networks. Because for training purposes these functions need to be differentiable, these translations are called differentiable logics (DL). This raises several research questions. What kind of differentiable logics are possible? What difference does a specific choice of DL make in the context of continuous verification? What are the desirable criteria for a DL viewed from the point of view of the resulting loss function? In this extended abstract we will discuss and answer these questions.

cs.AI↗
arXiv2022

Neural Network Robustness as a Verification Property: A Principled Case Study

Neural networks are very successful at detecting patterns in noisy data, and have become the technology of choice in many fields. However, their usefulness is hampered by their susceptibility to adversarial attacks. Recently, many methods for measuring and improving a network's robustness to adversarial perturbations have been proposed, and this growing body of research has given rise to numerous explicit or implicit notions of robustness. Connections between these notions are often subtle, and a systematic comparison between them is missing in the literature. In this paper we begin addressing this gap, by setting up general principles for the empirical analysis and evaluation of a network's robustness as a mathematical property - during the network's training phase, its verification, and after its deployment. We then apply these principles and conduct a case study that showcases the practical benefits of our general approach.

cs.LG↗
arXiv2022

Why Robust Natural Language Understanding is a Challenge

With the proliferation of Deep Machine Learning into real-life applications, a particular property of this technology has been brought to attention: robustness Neural Networks notoriously present low robustness and can be highly sensitive to small input perturbations. Recently, many methods for verifying networks' general properties of robustness have been proposed, but they are mostly applied in Computer Vision. In this paper we propose a Verification specification for Natural Language Understanding classification based on larger regions of interest, and we discuss the challenges of such task. We observe that, although the data is almost linearly separable, the verifier struggles to output positive results and we explain the problems and implications.

cs.CL↗
arXiv2022

Coinduction in Uniform: Foundations for Corecursive Proof Search with Horn Clauses

We establish proof-theoretic, constructive and coalgebraic foundations for proof search in coinductive Horn clause theories. Operational semantics of coinductive Horn clause resolution is cast in terms of coinductive uniform proofs; its constructive content is exposed via soundness relative to an intuitionistic first-order logic with recursion controlled by the later modality; and soundness of both proof systems is proven relative to a novel coalgebraic description of complete Herbrand models.

cs.LO↗
arXiv2021

Actions You Can Handle: Dependent Types for AI Plans

Verification of AI is a challenge that has engineering, algorithmic and programming language components. For example, AI planners are deployed to model actions of autonomous agents. They comprise a number of searching algorithms that, given a set of specified properties, find a sequence of actions that satisfy these properties. Although AI planners are mature tools from the algorithmic and engineering points of view, they have limitations as programming languages. Decidable and efficient automated search entails restrictions on the syntax of the language, prohibiting use of higher-order properties or recursion. This paper proposes a methodology for embedding plans produced by AI planners into dependently-typed language Agda, which enables users to reason about and verify more general and abstract properties of plans, and also provides a more holistic programming language infrastructure for modelling plan execution.

cs.AI↗
arXiv2020

Proof-Carrying Plans: a Resource Logic for AI Planning

Recent trends in AI verification and Explainable AI have raised the question of whether AI planning techniques can be verified. In this paper, we present a novel resource logic, the Proof Carrying Plans (PCP) logic that can be used to verify plans produced by AI planners. The PCP logic takes inspiration from existing resource logics (such as Linear logic and Separation logic) as well as Hoare logic when it comes to modelling states and resource-aware plan execution. It also capitalises on the Curry-Howard approach to logics, in its treatment of plans as functions and plan pre- and post-conditions as types. This paper presents two main results. From the theoretical perspective, we show that the PCP logic is sound relative to the standard possible world semantics used in AI planning. From the practical perspective, we present a complete Agda formalisation of the PCP logic and of its soundness proof. Moreover, we showcase the Curry-Howard, or functional, value of this implementation by supplementing it with the library that parses AI plans into Agda's proofs automatically. We provide evaluation of this library and the resulting Agda functions.

cs.LO↗
arXiv2020

The New Normal: We Cannot Eliminate Cuts in Coinductive Calculi, But We Can Explore Them

In sequent calculi, cut elimination is a property that guarantees that any provable formula can be proven analytically. For example, Gentzen's classical and intuitionistic calculi LK and LJ enjoy cut elimination. The property is less studied in coinductive extensions of sequent calculi. In this paper, we use coinductive Horn clause theories to show that cut is not eliminable in a coinductive extension of LJ, a system we call CLJ. We derive two further practical results from this study. We show that CoLP by Gupta et al. gives rise to cut-free proofs in CLJ with fixpoint terms, and we formulate and implement a novel method of coinductive theory exploration that provides several heuristics for discovery of cut formulae in CLJ.

cs.LO↗
arXiv2019

Neural Network Verification for the Masses (of AI graduates)

Rapid development of AI applications has stimulated demand for, and has given rise to, the rapidly growing number and diversity of AI MSc degrees. AI and Robotics research communities, industries and students are becoming increasingly aware of the problems caused by unsafe or insecure AI applications. Among them, perhaps the most famous example is vulnerability of deep neural networks to ``adversarial attacks''. Owing to wide-spread use of neural networks in all areas of AI, this problem is seen as particularly acute and pervasive. Despite of the growing number of research papers about safety and security vulnerabilities of AI applications, there is a noticeable shortage of accessible tools, methods and teaching materials for incorporating verification into AI programs. LAIV -- the Lab for AI and Verification -- is a newly opened research lab at Heriot-Watt university that engages AI and Robotics MSc students in verification projects, as part of their MSc dissertation work. In this paper, we will report on successes and unexpected difficulties LAIV faces, many of which arise from limitations of existing programming languages used for verification. We will discuss future directions for incorporating verification into AI degrees.

cs.AI↗
arXiv2019

Coinductive Uniform Proofs

Coinduction occurs in two guises in Horn clause logic: in proofs of circular properties and relations, and in proofs involving construction of infinite data. Both instances of coinductive reasoning appeared in the literature before, but a systematic analysis of these two kinds of proofs and of their relation was lacking. We propose a general proof-theoretic framework for handling both kinds of coinduction arising in Horn clause logic. To this aim, we propose a coinductive extension of Miller et al framework of uniform proofs and prove its soundness relative to coinductive models of Horn clause logic.

cs.LO↗
arXiv2017

Structural Resolution for Abstract Compilation of Object-Oriented Languages

We propose abstract compilation for precise static type analysis of object-oriented languages based on coinductive logic programming. Source code is translated to a logic program, then type-checking and inference problems amount to queries to be solved with respect to the resulting logic program. We exploit a coinductive semantics to deal with infinite terms and proofs produced by recursive types and methods. Thanks to the recent notion of structural resolution for coinductive logic programming, we are able to infer very precise type information, including a class of irrational recursive types causing non-termination for previously considered coinductive semantics. We also show how to transform logic programs to make them satisfy the preconditions for the operational semantics of structural resolution, and we prove this step does not affect the semantics of the logic program.

cs.PL↗
arXiv2017

Productive Corecursion in Logic Programming

Logic Programming is a Turing complete language. As a consequence, designing algorithms that decide termination and non-termination of programs or decide inductive/coinductive soundness of formulae is a challenging task. For example, the existing state-of-the-art algorithms can only semi-decide coinductive soundness of queries in logic programming for regular formulae. Another, less famous, but equally fundamental and important undecidable property is productivity. If a derivation is infinite and coinductively sound, we may ask whether the computed answer it determines actually computes an infinite formula. If it does, the infinite computation is productive. This intuition was first expressed under the name of computations at infinity in the 80s. In modern days of the Internet and stream processing, its importance lies in connection to infinite data structure processing. Recently, an algorithm was presented that semi-decides a weaker property -- of productivity of logic programs. A logic program is productive if it can give rise to productive derivations. In this paper we strengthen these recent results. We propose a method that semi-decides productivity of individual derivations for regular formulae. Thus we at last give an algorithmic counterpart to the notion of productivity of derivations in logic programming. This is the first algorithmic solution to the problem since it was raised more than 30 years ago. We also present an implementation of this algorithm.

cs.LO↗
arXiv2017

Proof Mining with Dependent Types

Several approaches exist to data-mining big corpora of formal proofs. Some of these approaches are based on statistical machine learning, and some -- on theory exploration. However, most are developed for either untyped or simply-typed theorem provers. In this paper, we present a method that combines statistical data mining and theory exploration in order to analyse and automate proofs in dependently typed language of Coq.

cs.PL↗
arXiv2017

Structural Resolution: a Framework for Coinductive Proof Search and Proof Construction in Horn Clause Logic

Logic programming (LP) is a programming language based on first-order Horn clause logic that uses SLD-resolution as a semi-decision procedure. Finite SLD-computations are inductively sound and complete with respect to least Herbrand models of logic programs. Dually, the corecursive approach to SLD-resolution views infinite SLD-computations as successively approximating infinite terms contained in programs' greatest complete Herbrand models. State-of-the-art algorithms implementing corecursion in LP are based on loop detection. However, such algorithms support inference of logical entailment only for rational terms, and they do not account for the important property of productivity in infinite SLD-computations. Loop detection thus lags behind coinductive methods in interactive theorem proving (ITP) and term-rewriting systems (TRS). Structural resolution is a newly proposed alternative to SLD-resolution that makes it possible to define and semi-decide a notion of productivity appropriate to LP. In this paper, we prove soundness of structural resolution relative to Herbrand model semantics for productive inductive, coinductive, and mixed inductive-coinductive logic programs. We introduce two algorithms that support coinductive proof search for infinite productive terms. One algorithm combines the method of loop detection with productive structural resolution, thus guaranteeing productivity of coinductive proofs for infinite rational terms. The other allows to make lazy sound observations of fragments of infinite irrational productive terms. This puts coinductive methods in LP on par with productivity-based observational approaches to coinduction in ITP and TRS.

cs.LO↗
arXiv2016

CoALP-Ty'16

This volume constitutes the pre-proceedings of the Workshop on Coalgebra, Horn Clause Logic Programming and Types (CoALP-Ty'16), held on 28--29 November 2016 in Edinburgh as a mark of the end of the EPSRC Grant Coalgebraic Logic Programming for Type Inference, by E. Komendantskaya and J. Power. This volume consists of extended abstracts describing current research in the following areas: Semantics: Lawvere theories and Coalgebra in Logic and Functional Programming Programming languages: Horn Clause Logic for Type Inference in Functional Languages and Beyond After discussion at the workshop authors of the extended abstracts will be invited to submit a full paper to go through a second round of refereeing and selection for the formal proceedings.

cs.LO↗
arXiv2016

Coinductive Soundness of Corecursive Type Class Resolution

Horn clauses and first-order resolution are commonly used to implement type classes in Haskell. Several corecursive extensions to type class resolution have recently been proposed, with the goal of allowing (co)recursive dictionary construction where resolution does not termi- nate. This paper shows, for the first time, that corecursive type class resolution and its extensions are coinductively sound with respect to the greatest Herbrand models of logic programs and that they are induc- tively unsound with respect to the least Herbrand models. We establish incompleteness results for various fragments of the proof system.

cs.PL↗