Searcharxiv⌕ Search

arXiv subjects

Samson Abramsky

Publications and source records attributed to Samson Abramsky.

At least 55 records · Page 3Linked to original sources
arXiv2017

The pebbling comonad in finite model theory

Pebble games are a powerful tool in the study of finite model theory, constraint satisfaction and database theory. Monads and comonads are basic notions of category theory which are widely used in semantics of computation and in modern functional programming. We show that existential k-pebble games have a natural comonadic formulation. Winning strategies for Duplicator in the k-pebble game for structures A and B are equivalent to morphisms from A to B in the coKleisli category for this comonad. This leads on to comonadic characterisations of a number of central concepts in Finite Model Theory: - Isomorphism in the co-Kleisli category characterises elementary equivalence in the k-variable logic with counting quantifiers. - Symmetric games corresponding to equivalence in full k-variable logic are also characterized. - The treewidth of a structure A is characterised in terms of its coalgebra number: the least k for which there is a coalgebra structure on A for the k-pebbling comonad. - Co-Kleisli morphisms are used to characterize strong consistency, and to give an account of a Cai-Fürer-Immerman construction. - The k-pebbling comonad is also used to give semantics to a novel modal operator. These results lay the basis for some new and promising connections between two areas within logic in computer science which have largely been disjoint: (1) finite and algorithmic model theory, and (2) semantics and categorical structures of computation.

cs.LO↗
arXiv2017

Contextuality, Cohomology and Paradox

Contextuality is a key feature of quantum mechanics that provides an important non-classical resource for quantum information and computation. Abramsky and Brandenburger used sheaf theory to give a general treatment of contextuality in quantum theory [New Journal of Physics 13 (2011) 113036]. However, contextual phenomena are found in other fields as well, for example database theory. In this paper, we shall develop this unified view of contextuality. We provide two main contributions: firstly, we expose a remarkable connection between contexuality and logical paradoxes; secondly, we show that an important class of contextuality arguments has a topological origin. More specifically, we show that "All-vs-Nothing" proofs of contextuality are witnessed by cohomological obstructions.

quant-ph↗
arXiv2016

Anytime Inference in Valuation Algebras

Anytime inference is inference performed incrementally, with the accuracy of the inference being controlled by a tunable parameter, usually time. Such anytime inference algorithms are also usually interruptible, gradually converging to the exact inference value until terminated. While anytime inference algorithms for specific domains like probability potentials exist in the literature, our objective in this article is to obtain an anytime inference algorithm which is sufficiently generic to cover a wide range of domains. For this we utilise the theory of generic inference as a basis for constructing an anytime inference algorithm, and in particular, extending work done on ordered valuation algebras. The novel contribution of this work is the construction of anytime algorithms in a generic framework, which automatically gives us instantiations in various useful domains. We also show how to apply this generic framework for anytime inference in semiring induced valuation algebras, an important subclass of valuation algebras, which includes instances like probability potentials, disjunctive normal forms and distributive lattices. Keywords: Approximation; Anytime algorithms; Resource-bounded computation; Generic inference; Valuation algebras; Local computation; Binary join trees.

cs.AI↗
arXiv2016

Information, Processes and Games

We survey the prospects for an Information Dynamics which can serve as the basis for a fundamental theory of information, incorporating qualitative and structural as well as quantitative aspects. We motivate our discussion with some basic conceptual puzzles: how can information increase in computation, and what is it that we are actually computing in general? Then we survey a number of the theories which have been developed within Computer Science, as partial exemplifications of the kind of fundamental theory which we seek: including Domain Theory, Dynamic Logic, and Process Algebra. We look at recent work showing new ways of combining quantitative and qualitative theories of information, as embodied respectively by Domain Theory and Shannon Information Theory. Then we look at Game Semantics and Geometry of Interaction, as examples of dynamic models of logic and computation in which information flow and interaction are made central and explicit. We conclude by looking briefly at some key issues for future progress.

cs.LO↗
arXiv2016

Possibilities Determine the Combinatorial Structure of Probability Polytopes

We study the set of no-signalling empirical models on a measurement scenario, and show that the combinatorial structure of the no-signalling polytope is completely determined by the possibilistic information given by the support of the models. This is a special case of a general result which applies to all polytopes presented in a standard form, given by linear equations together with non-negativity constraints on the variables.

quant-ph↗
arXiv2015

Process Realizability

We develop a notion of realizability for Classical Linear Logic based on a concurrent process calculus.

cs.LO↗
arXiv2015

DEMONIC programming: a computational language for single-particle equilibrium thermodynamics, and its formal semantics

Maxwell's Demon, 'a being whose faculties are so sharpened that he can follow every molecule in its course', has been the centre of much debate about its abilities to violate the second law of thermodynamics. Landauer's hypothesis, that the Demon must erase its memory and incur a thermodynamic cost, has become the standard response to Maxwell's dilemma, and its implications for the thermodynamics of computation reach into many areas of quantum and classical computing. It remains, however, still a hypothesis. Debate has often centred around simple toy models of a single particle in a box. Despite their simplicity, the ability of these systems to accurately represent thermodynamics (specifically to satisfy the second law) and whether or not they display Landauer Erasure, has been a matter of ongoing argument. The recent Norton-Ladyman controversy is one such example. In this paper we introduce a programming language to describe these simple thermodynamic processes, and give a formal operational semantics and program logic as a basis for formal reasoning about thermodynamic systems. We formalise the basic single-particle operations as statements in the language, and then show that the second law must be satisfied by any composition of these basic operations. This is done by finding a computational invariant of the system. We show, furthermore, that this invariant requires an erasure cost to exist within the system, equal to kTln2 for a bit of information: Landauer Erasure becomes a theorem of the formal system. The Norton-Ladyman controversy can therefore be resolved in a rigorous fashion, and moreover the formalism we introduce gives a set of reasoning tools for further analysis of Landauer erasure, which are provably consistent with the second law of thermodynamics.

cs.LO↗
arXiv2015

Games for Dependent Types

We present a model of dependent type theory (DTT) with Pi-, 1-, Sigma- and intensional Id-types, which is based on a slight variation of the category of AJM-games and history-free winning strategies. The model satisfies Streicher's criteria of intensionality and refutes function extensionality. The principle of uniqueness of identity proofs is satisfied. We show it contains a submodel as a full subcategory which gives a faithful model of DTT with Pi-, 1-, Sigma- and intensional Id-types and, additionally, finite inductive type families. This smaller model is fully (and faithfully) complete with respect to the syntax at the type hierarchy built without Id-types, as well as at the class of types where we allow for one strictly positive occurrence of an Id-type. Definability for the full type hierarchy with Id-types remains to be investigated.

cs.LO↗
arXiv2015

Hardy is (almost) everywhere: nonlocality without inequalities for almost all entangled multipartite states

We show that all $n$-qubit entangled states, with the exception of tensor products of single-qubit and bipartite maximally-entangled states, admit Hardy-type proofs of non-locality without inequalities or probabilities. More precisely, we show that for all such states, there are local, one-qubit observables such that the resulting probability tables are logically contextual in the sense of Abramsky and Brandenburger, this being the general form of the Hardy-type property. Moreover, our proof is constructive: given a state, we show how to produce the witnessing local observables. In fact, we give an algorithm to do this. Although the algorithm is reasonably straightforward, its proof of correctness is non-trivial. A further striking feature is that we show that $n+2$ local observables suffice to witness the logical contextuality of any $n$-qubit state: two each for two for the parties, and one each for the remaining $n-2$ parties.

quant-ph↗
arXiv2014

No-Signalling Is Equivalent To Free Choice of Measurements

No-Signalling is a fundamental constraint on the probabilistic predictions made by physical theories. It is usually justified in terms of the constraints imposed by special relativity. However, this justification is not as clear-cut as is usually supposed. We shall give a different perspective on this condition by showing an equivalence between No-Signalling and Lambda Independence, or "free choice of measurements", a condition on hidden-variable theories which is needed to make no-go theorems such as Bell's theorem non-trivial. More precisely, we shall show that a probability table describing measurement outcomes is No-Signalling if and only if it can be realized by a Lambda-Independent hidden-variable theory of a particular canonical form, in which the hidden variables correspond to non-contextual deterministic predictions of measurement outcomes. The key proviso which avoids contradiction with Bell's theorem is that we consider hidden-variable theories with signed probability measures over the hidden variables - i.e. negative probabilities. Negative probabilities have often been discussed in the literature on quantum mechanics. We use a result proved previously in "The Sheaf-theoretic Structure of Locality and Contextuality" by Abramsky and Brandenburger, which shows that they give rise to, and indeed characterize, the entire class of No-Signalling behaviours. In the present paper, we put this result in a broader context, which reveals the surprising consequence that the No-Signalling condition is equivalent to the apparently completely different notion of free choice of measurements.

quant-ph↗
arXiv2014

A classification of multipartite states by degree of non-locality

We propose a novel form of classification of multipartite states, in terms of the maximum degree of non-locality they can exhibit under any choice of local observables. This uses the hierarchy of notions previously introduced by Abramsky and Brandenburger: strong contextuality, logical contextuality, and probabilistic contextuality. We study n-qubit pure states. We conjecture that for more than 2 parties, all entangled states are logically contextual. We prove a number of results in support of this conjecture: (1) We show that all permutation-symmetric states are logically non-local. (2) We study the class of balanced states with functional dependencies. These states are described by Boolean functions and have a rich structure, allowing a detailed analysis, which again confirms the conjecture in this case.

quant-ph↗
arXiv2014

Two Puzzles About Computation

The purpose of this note is to raise two different questions, which are rarely if ever considered, and to which, it seems, we lack convincing, systematic answers. These questions can be posed as: - Why do we compute? - What do we compute? The point is not so much that we have no answers to these puzzles, as that we have no established body of theory which gives satisfying, systematic answers, as part of a broader understanding. By raising these questions, we hope to stimulate some thinking in this direction.

cs.LO↗
arXiv2014

Semantic Unification A sheaf theoretic approach to natural language

Language is contextual and sheaf theory provides a high level mathematical framework to model contextuality. We show how sheaf theory can model the contextual nature of natural language and how gluing can be used to provide a global semantics for a discourse by putting together the local logical semantics of each sentence within the discourse. We introduce a presheaf structure corresponding to a basic form of Discourse Representation Structures. Within this setting, we formulate a notion of semantic unification --- gluing meanings of parts of a discourse into a coherent whole --- as a form of sheaf-theoretic gluing. We illustrate this idea with a number of examples where it can used to represent resolutions of anaphoric references. We also discuss multivalued gluing, described using a distributions functor, which can be used to represent situations where multiple gluings are possible, and where we may need to rank them using quantitative measures. Dedicated to Jim Lambek on the occasion of his 90th birthday.

cs.CL↗
arXiv2014

Axioms for Definability and Full Completeness

Axioms are presented which encapsulate the properties satisfied by categories of games which form the basis of results on full abstraction for PCF and other programming languages, and on full completeness for various logics and type theories. Axioms are presented on models of PCF from which full abstraction can be proved. These axioms have been distilled from recent results on definability and full abstraction of game semantics for a number of programming languages. Full completeness for pure simply-typed $λ$-calculus is also axiomatized.

cs.LO↗
arXiv2014

Game Semantics for Access Control

We introduce a semantic approach to the study of logics for access control and dependency analysis, based on Game Semantics. We use a variant of AJM games with explicit justification (but without pointers). Based on this, we give a simple and intuitive model of the information flow constraints underlying access control. This is used to give strikingly simple proofs of \emph{non-interference theorems} in robust, semantic versions.

cs.LO↗