https://arxiv.org/api/BouPABc+j6YbgEY316hD9gpare0 2026-04-19T11:51:53Z 3075 690 15 http://arxiv.org/abs/2403.16269v1 2024-03-24T19:16:37Z

This article serves as a preliminary introduction to the design of a new, open-source applied and computational category theory framework, named Categorica, built on top of the Wolfram Language. Categorica allows one to configure and manipulate abstract quivers, categories, groupoids, diagrams, functors and natural transformations, and to perform a vast array of automated abstract algebraic computations using (arbitrary combinations of) the above structures; to manipulate and abstractly reason about arbitrary universal properties, including products, coproducts, pullbacks, pushouts, limits and colimits; and to manipulate, visualize and compute with strict (symmetric) monoidal categories, including full support for automated string diagram rewriting and diagrammatic theorem-proving. In so doing, Categorica combines the capabilities of an abstract computer algebra framework (thus allowing one to compute directly with epimorphisms, monomorphisms, retractions, sections, spans, cospans, fibrations, etc.) with those of a powerful automated theorem-proving system (thus allowing one to convert universal properties and other abstract constructions into (higher-order) equational logic statements that can be reasoned about and proved using standard automated theorem-proving methods, as well as to prove category-theoretic statements directly using purely diagrammatic methods). In this first of two articles introducing the design of the framework, we shall focus principally upon its handling of quivers, categories, diagrams, groupoids, functors and natural transformations, including demonstrations of both its algebraic manipulation and theorem-proving capabilities in each case.

2024-03-24T19:16:37Z 71 pages, 29 figures Jonathan Gorard http://arxiv.org/abs/2403.13218v1 2024-03-20T00:37:19Z

Vector Symbolic Architectures (VSAs) have emerged as a novel framework for enabling interpretable machine learning algorithms equipped with the ability to reason and explain their decision processes. The basic idea is to represent discrete information through high dimensional random vectors. Complex data structures can be built up with operations over vectors such as the "binding" operation involving element-wise vector multiplication, which associates data together. The reverse task of decomposing the associated elements is a combinatorially hard task, with an exponentially large search space. The main algorithm for performing this search is the resonator network, inspired by Hopfield network-based memory search operations. In this work, we introduce a new variant of the resonator network, based on self-attention based update rules in the iterative search problem. This update rule, based on the Hopfield network with log-sum-exp energy function and norm-bounded states, is shown to substantially improve the performance and rate of convergence. As a result, our algorithm enables a larger capacity for associative memory, enabling applications in many tasks like perception based pattern recognition, scene decomposition, and object reasoning. We substantiate our algorithm with a thorough evaluation and comparisons to baselines.

2024-03-20T00:37:19Z Calvin Yeung Prathyush Poduval Mohsen Imani http://arxiv.org/abs/2302.05954v2 2024-03-19T14:36:11Z

This paper presents an up-to-date and refined version of the SCL calculus for first-order logic without equality. The refinement mainly consists of the following two parts: First, we incorporate a stronger notion of regularity into SCL(FOL). Our regularity definition is adapted from the SCL(T) calculus. This adapted definition guarantees non-redundant clause learning during a run of SCL. However, in contrast to the original presentation, it does not require exhaustive propagation. Second, we introduce trail and model bounding to achieve termination guarantees. In previous versions, no termination guarantees about SCL were achieved. Last, we give rigorous proofs for soundness, completeness and clause learning guarantees of SCL(FOL) and put SCL(FOL) into context of existing first-order calculi.

2023-02-12T16:52:56Z Martin Bromberger Simon Schwarz Christoph Weidenbach http://arxiv.org/abs/2403.12153v1 2024-03-18T18:09:47Z

We present alternative approaches to routing and scheduling in Answer Set Programming (ASP), and explore them in the context of Multi-agent Path Finding. The idea is to capture the flow of time in terms of partial orders rather than time steps attached to actions and fluents. This also abolishes the need for fixed upper bounds on the length of plans. The trade-off for this avoidance is that (parts of) temporal trajectories must be acyclic, since multiple occurrences of the same action or fluent cannot be distinguished anymore. While this approach provides an interesting alternative for modeling routing, it is without alternative for scheduling since fine-grained timings cannot be represented in ASP in a feasible way. This is different for partial orders that can be efficiently handled by external means such as acyclicity and difference constraints. We formally elaborate upon this idea and present several resulting ASP encodings. Finally, we demonstrate their effectiveness via an empirical analysis.

2024-03-18T18:09:47Z Roland Kaminski Torsten Schaub Tran Cao Son Jiří Švancara Philipp Wanko http://arxiv.org/abs/2312.16960v2 2024-03-16T13:36:17Z

This study proposes the "adaptive flip graph algorithm", which combines adaptive searches with the flip graph algorithm for finding fast and efficient methods for matrix multiplication. The adaptive flip graph algorithm addresses the inherent limitations of exploration and inefficient search encountered in the original flip graph algorithm, particularly when dealing with large matrix multiplication. For the limitation of exploration, the proposed algorithm adaptively transitions over the flip graph, introducing a flexibility that does not strictly reduce the number of multiplications. Concerning the issue of inefficient search in large instances, the proposed algorithm adaptively constraints the search range instead of relying on a completely random search, facilitating more effective exploration. Numerical experimental results demonstrate the effectiveness of the adaptive flip graph algorithm, showing a reduction in the number of multiplications for a $4\times 5$ matrix multiplied by a $5\times 5$ matrix from $76$ to $73$, and that from $95$ to $94$ for a $5 \times 5$ matrix multiplied by another $5\times 5$ matrix. These results are obtained in characteristic two.

2023-12-28T11:05:55Z 7 pages, 2 figures Yamato Arai Yuma Ichikawa Koji Hukushima http://arxiv.org/abs/2309.02427v3 2024-03-15T15:44:11Z

Recent efforts have augmented large language models (LLMs) with external resources (e.g., the Internet) or internal control flows (e.g., prompt chaining) for tasks requiring grounding or reasoning, leading to a new class of language agents. While these agents have achieved substantial empirical success, we lack a systematic framework to organize existing agents and plan future developments. In this paper, we draw on the rich history of cognitive science and symbolic artificial intelligence to propose Cognitive Architectures for Language Agents (CoALA). CoALA describes a language agent with modular memory components, a structured action space to interact with internal memory and external environments, and a generalized decision-making process to choose actions. We use CoALA to retrospectively survey and organize a large body of recent work, and prospectively identify actionable directions towards more capable agents. Taken together, CoALA contextualizes today's language agents within the broader history of AI and outlines a path towards language-based general intelligence.

2023-09-05T17:56:20Z v3 is TMLR camera ready version. 19 pages of main content, 5 figures. The first two authors contributed equally, order decided by coin flip. A CoALA-based repo of recent work on language agents: https://github.com/ysymyth/awesome-language-agents Theodore R. Sumers Shunyu Yao Karthik Narasimhan Thomas L. Griffiths http://arxiv.org/abs/2403.10260v1 2024-03-15T12:48:41Z

Differential-algebraic equations (DAEs) have been used in modeling various dynamical systems in science and engineering. Several preprocessing methods for DAEs, such as consistent initialization and index reduction, use structural information on DAEs. Unfortunately, these methods may fail when the system Jacobian, which is a functional matrix, derived from the DAE is singular. To transform a DAE with a singular system Jacobian into a nonsingular system, several regularization methods have been proposed. Most of all existing regularization methods rely on symbolic computation to eliminate the system Jacobian for finding a certificate of singularity, resulting in much computational time. Iwata--Oki--Takamatsu (2019) proposed a method (IOT-method) to find a certificate without symbolic computations. The IOT method approximates the system Jacobian by a simpler symbolic matrix, called a layered mixed matrix, which admits a fast combinatorial algorithm for singularity testing. However, it often overlooks the singularity of the system Jacobian since the approximation largely discards algebraic relationships among entries in the original system Jacobian. In this study, we propose a new regularization method extending the idea of the IOT method. Instead of layered mixed matrices, our method approximates the system Jacobian by more expressive symbolic matrices, called rank-1 coefficient mixed (1CM) matrices. This makes our method more widely applicable. We give a fast combinatorial algorithm for finding a singularity certificate of 1CM-matrices, which is free from symbolic elimination. Our method is also advantageous in that it globally preserves the solution set to the DAE. Through numerical experiments, we confirmed that our method runs fast for large-scale DAEs from real instances.

2024-03-15T12:48:41Z Taihei Oki Yujin Song http://arxiv.org/abs/2403.09145v1 2024-03-14T07:47:02Z

We study the computational complexity of counting constraint satisfaction problems (#CSPs) whose constraints assign complex numbers to Boolean inputs when the corresponding constraint hypergraphs are acyclic. These problems are called acyclic #CSPs or succinctly, #ACSPs. We wish to determine the computational complexity of all such #ACSPs when arbitrary unary constraints are freely available. Depending on whether we further allow or disallow the free use of the specific constraint XOR (binary disequality), we present two complexity classifications of the #ACSPs according to the types of constraints used for the problems. When XOR is freely available, we first obtain a complete dichotomy classification. On the contrary, when XOR is not available for free, we then obtain a trichotomy classification. To deal with an acyclic nature of constraints in those classifications, we develop a new technical tool called acyclic-T-constructibility or AT-constructibility, and we exploit it to analyze a complexity upper bound of each #ACSPs.

2024-03-14T07:47:02Z (A4, 10pt, 17 pages) An extended abstract of this current article is scheduled to appear in the Proceedings of the 12th Computing Conference, London, UK, July 11--12, 2024, Lecture Notes in Networks and Systems, Springer-Verlag, 2024 Tomoyuki Yamakami http://arxiv.org/abs/2403.06542v2 2024-03-13T08:52:18Z

The solutions of the equation f^{ (p--1) }+ f^p = h^p in the unknown function f overan algebraic function field of characteristic p are very closely linked to the structure and fac-torisations of linear differential operators with coefficients in function fields of characteristic p.However, while being able to solve this equation over general algebraic function fields is necessaryeven for operators with rational coefficients, no general resolution method has been developed.We present an algorithm for testing the existence of solutions in polynomial time in the ``size''of h and an algorithm based on the computation of Riemann-Roch spaces and the selection ofelements in the divisor class group, for computing solutions of size polynomial in the ``size'' of hin polynomial time in the size of h and linear in the characteristic p, and discuss its applicationsto the factorisation of linear differential operators in positive characteristic p.

2024-03-11T09:36:54Z Raphaël Pagès UB, IMB, MATHEXP http://arxiv.org/abs/2403.07995v1 2024-03-12T18:03:08Z

Modelling musical structure is vital yet challenging for artificial intelligence systems that generate symbolic music compositions. This literature review dissects the evolution of techniques for incorporating coherent structure, from symbolic approaches to foundational and transformative deep learning methods that harness the power of computation and data across a wide variety of training paradigms. In the later stages, we review an emerging technique which we refer to as "sub-task decomposition" that involves decomposing music generation into separate high-level structural planning and content creation stages. Such systems incorporate some form of musical knowledge or neuro-symbolic methods by extracting melodic skeletons or structural templates to guide the generation. Progress is evident in capturing motifs and repetitions across all three eras reviewed, yet modelling the nuanced development of themes across extended compositions in the style of human composers remains difficult. We outline several key future directions to realize the synergistic benefits of combining approaches from all eras examined.

2024-03-12T18:03:08Z Accepted to 13th International Conference on Artificial Intelligence in Music, Sound, Art and Design (EvoMUSART) 2024 Keshav Bhandari Simon Colton http://arxiv.org/abs/2203.06970v6 2024-03-12T13:45:27Z

We explore algorithmic aspects of a simply transitive commutative group action coming from the class field theory of imaginary hyperelliptic function fields. Namely, the Jacobian of an imaginary hyperelliptic curve defined over $\mathbb F_q$ acts on a subset of isomorphism classes of Drinfeld modules. We describe an algorithm to compute the group action efficiently. This is a function field analog of the Couveignes-Rostovtsev-Stolbunov group action. We report on an explicit computation done with our proof-of-concept C++/NTL implementation; it took a fraction of a second on a standard computer. We prove that the problem of inverting the group action reduces to the problem of finding isogenies of fixed $τ$-degree between Drinfeld $\mathbb F_q[X]$-modules, which is solvable in polynomial time thanks to an algorithm by Wesolowski. We give asymptotic complexity bounds for all algorithms presented in this paper.

2022-03-14T10:11:35Z This paper is a rewrite of arXiv:2203.06970v2. It takes into account the recent attack of Wesolowski on the cryptographic applications (https://eprint.iacr.org/2022/438). We removed cryptographic applications, and the introduction and experimental results have been widely rewritten. Complexity results have been added Antoine Leudière Pierre-Jean Spaenlehauer http://arxiv.org/abs/2310.19699v3 2024-03-07T20:51:18Z

The Logical Execution Time (LET) programming model has recently received considerable attention, particularly because of its timing and dataflow determinism. In LET, task computation appears always to take the same amount of time (called the task's LET interval), and the task reads (resp. writes) at the beginning (resp. end) of the interval. Compared to other communication mechanisms, such as implicit communication and Dynamic Buffer Protocol (DBP), LET performs worse on many metrics, such as end-to-end latency (including reaction time and data age) and time disparity jitter. Compared with the default LET setting, the flexible LET (fLET) model shrinks the LET interval while still guaranteeing schedulability by introducing the virtual offset to defer the read operation and using the virtual deadline to move up the write operation. Therefore, fLET has the potential to significantly improve the end-to-end timing performance while keeping the benefits of deterministic behavior on timing and dataflow. To fully realize the potential of fLET, we consider the problem of optimizing the assignments of its virtual offsets and deadlines. We propose new abstractions to describe the task communication pattern and new optimization algorithms to explore the solution space efficiently. The algorithms leverage the linearizability of communication patterns and utilize symbolic operations to achieve efficient optimization while providing a theoretical guarantee. The framework supports optimizing multiple performance metrics and guarantees bounded suboptimality when optimizing end-to-end latency. Experimental results show that our optimization algorithms improve upon the default LET and its existing extensions and significantly outperform implicit communication and DBP in terms of various metrics, such as end-to-end latency, time disparity, and its jitter.

2023-10-30T16:21:49Z accepted in RTAS'24 Sen Wang Dong Li Ashrarul H. Sifat Shao-Yu Huang Xuanliang Deng Changhee Jung Ryan Williams Haibo Zeng http://arxiv.org/abs/2403.04017v1 2024-03-06T19:59:17Z

Automated theorem provers and formal proof assistants are general reasoning systems that are in theory capable of proving arbitrarily hard theorems, thus solving arbitrary problems reducible to mathematics and logical reasoning. In practice, such systems however face large combinatorial explosion, and therefore include many heuristics and choice points that considerably influence their performance. This is an opportunity for trained machine learning predictors, which can guide the work of such reasoning systems. Conversely, deductive search supported by the notion of logically valid proof allows one to train machine learning systems on large reasoning corpora. Such bodies of proof are usually correct by construction and when combined with more and more precise trained guidance they can be boostrapped into very large corpora, with increasingly long reasoning chains and possibly novel proof ideas. In this paper we provide an overview of several automated reasoning and theorem proving domains and the learning and AI methods that have been so far developed for them. These include premise selection, proof guidance in several settings, AI systems and feedback loops iterating between reasoning and learning, and symbolic classification problems.

2024-03-06T19:59:17Z Lasse Blaauwbroek David Cerna Thibault Gauthier Jan Jakubův Cezary Kaliszyk Martin Suda Josef Urban 10.1007/978-3-031-61716-4_4 http://arxiv.org/abs/2403.03712v1 2024-03-06T13:54:57Z

We present a first-order theorem proving framework for establishing the correctness of functional programs implementing sorting algorithms with recursive data structures. We formalize the semantics of recursive programs in many-sorted first-order logic and integrate sortedness/permutation properties within our first-order formalization. Rather than focusing on sorting lists of elements of specific first-order theories, such as integer arithmetic, our list formalization relies on a sort parameter abstracting (arithmetic) theories and hence concrete sorts. We formalize the permutation property of lists in first-order logic so that we automatically prove verification conditions of such algorithms purely by superpositon-based first-order reasoning. Doing so, we adjust recent efforts for automating inducion in saturation. We advocate a compositional approach for automating proofs by induction required to verify functional programs implementing and preserving sorting and permutation properties over parameterized list structures. Our work turns saturation-based first-order theorem proving into an automated verification engine by (i) guiding automated inductive reasoning with manual proof splits and (ii) fully automating inductive reasoning in saturation. We showcase the applicability of our framework over recursive sorting algorithms, including Mergesort and Quicksort.

2024-03-06T13:54:57Z Pamina Georgiou Márton Hajdu Laura Kovács http://arxiv.org/abs/2310.08047v2 2024-03-05T16:06:09Z

We solve the so far open problem of constructing all spatial rational curves with rational arc length functions. More precisely, we present three different methods for this construction. The first method adapts a recent approach of (Kalkan et al. 2022) to rational PH curves and requires solving a modestly sized system of linear equations. The second constructs the curve by imposing zero-residue conditions, thus extending ideas of previous papers by (Farouki and Sakkalis 2019) and the authors themselves (Schröcker and Šír 2023). The third method generalizes the dual approach of (Pottmann 1995) from planar to spatial curves. The three methods share the same quaternion based representation in which not only the PH curve but also its arc length function are compactly expressed. We also present a new proof based on the quaternion polynomial factorization theory of the well known characterization of the Pythagorean quadruples.

2023-10-12T05:39:56Z Hans-Peter Schröcker Zbyněk Šìr