Neurosymbolic Artificial Intelligence - Volume Pre-press, issue Pre-press

Authors: Giunchiglia, Eleonora | Imrie, Fergus | van der Schaar, Mihaela | Lukasiewicz, Thomas

Article Type: Research Article

Abstract: In the recent years, machine learning has made great advancements that have been at the root of many breakthroughs in different application domains. However, it is still an open issue how to make them applicable to high-stakes or safety-critical application domains, as they can often be brittle and unreliable. In this paper, we argue that requirements definition and satisfaction can go a long way to make machine learning models even more fitting to the real world, especially in critical domains. To this end, we present two problems in which (i) requirements arise naturally, (ii) machine learning models are or can …be fruitfully deployed, and (iii) neglecting the requirements can have dramatic consequences. Our proposed pyramid development process integrates requirements specification into every stage of the machine learning pipeline, ensuring mutual influence between requirements and subsequent phases. Additionally, we explore the pivotal role of Neuro-symbolic AI in facilitating this integration, paving the way for more reliable and robust machine learning applications in critical domains. Through this approach, we aim to bridge the gap between theoretical advancements and practical implementations, ensuring machine learning’s safe and effective deployment in sensitive areas. Show more

Keywords: Safe AI, machine learning, requirements, machine learning operations, software engineering

DOI: 10.3233/NAI-240767

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-13, 2024

Authors: Šír, Gustav

Article Type: Research Article

Abstract: Neural-Symbolic Integration (NSI) aims to marry the principles of symbolic AI techniques, such as logical reasoning, with the learning capabilities of neural networks. In recent years, many systems have been proposed to address this integration in a seemingly efficient manner. However, from the computational perspective, this is in principle impossible to do. Specifically, some of the core symbolic problems are provably hard, hence a general NSI system necessarily needs to adopt this computational complexity, too. Many NSI methods try to circumvent this downside by inconspicuously dropping parts of the symbolic capabilities while mapping the problems into static tensor representations in …exchange for efficient deep learning acceleration. In this paper, we argue that the aim for a general NSI system, properly covering both the neural and symbolic paradigms, has important computational implications on the learning representations, the structure of the resulting computation graphs, and the underlying hardware and software stacks. Particularly, we explain how the currently prominent, tensor-based deep learning with static computation graphs is conceptually insufficient as a foundation for such general NSI, which we discuss in a wider context of established (statistical) relational and structured deep learning methods. Finally, we delve into the underlying hardware acceleration aspects and outline some promising computational directions toward fully expressive and efficient NSI. Show more

Keywords: Neural-symbolic integration, computation graphs, learning representations, relational learning, hardware acceleration, logical reasoning, dynamic deep learning, computational complexity

DOI: 10.3233/NAI-240672

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-12, 2024

Authors: Fanizzi, Nicola | d’Amato, Claudia

Article Type: Research Article

Abstract: The paper surveys ongoing research on hyperdimensional computing and vector symbolic architectures which represent an alternative approach to neural computing with various advantages and interesting specific properties: transparency, error tolerance, sustainability. In particular, it can be demonstrated that hyperdimensional patterns are well-suited for the encoding of complex knowledge structures. Consequently, the related architectures offer perspectives for the development of innovative neurosymbolic models with a closer correspondence to cognitive processes in human brains. We revisit the fundamentals of hyperdimensional representations and examine some recent applications of related methods for analogical reasoning and learning tasks, with a particular focus on knowledge graphs. …We then propose potential extensions and delineate prospective avenues for future investigations. Show more

Keywords: Hyperdimensional computing, vector symbolic architectures, analogical reasoning, similarity-based classification, logic axioms, knowledge graphs, embedding models

DOI: 10.3233/NAI-240675

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-15, 2024

Authors: Wagner, Benedikt J. | d’Avlia Garcez, Artur

Article Type: Research Article

Abstract: We propose neurosymbolic integration as an approach for AI alignment via concept-based model explanation. The aim is to offer AI systems the ability to learn from human revision but also assist humans at evaluating AI capabilities. The proposed method allows users and domain experts to learn about the data-driven decision making process of large neural network models and to impose a particular behaviour onto such models. The models are queried using a symbolic logic language that acts as a lingua franca between humans and model representations. Interaction with the user then confirms or rejects a revision of the model using …logical constraints that can be distilled back into the neural network. We illustrate the approach using the Logic Tensor Network framework alongside Concept Activation Vectors and apply it to Convolutional Neural Networks and the task of achieving quantitative fairness. Our results illustrate how the use of a logical language is able to provide users with a formalisation of the model’s decision making whilst allowing users to steer the model towards a given alignment constraint. Show more

Keywords: Neurosymbolic alignment, concept representation, knowledge revision, explainable AI

DOI: 10.3233/NAI-240729

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-12, 2024

Authors: Mileo, Alessandra

Article Type: Research Article

Abstract: Deep learning is being very successful in supporting humans in the interpretation of complex data (such as images and text) for critical decision tasks. However, it still remains difficult for human experts to understand how such results are achieved, due to the “black box” nature of the deep models used. In high-stake decision making scenarios such as the interpretation of medical imaging for diagnostics, such a lack of transparency still hinders the adoption of these techniques in practice. In this position paper we present a conceptual methodology for the design of a neuro-symbolic cycle to address the need for explainability …and confidence (including trust) of deep learning models when used to support human experts in high-stake decision making, and we discuss challenges and opportunities in the implementation of such cycle as well as its adoption in real world scenarios. We elaborate on the need to leverage the potential of hybrid artificial intelligence combining neural learning and symbolic reasoning in a human-centered approach to explainability. We advocate that the phases of such a cycle should include i) the extraction of knowledge from a trained network to represent and encode its behaviour, ii) the validation of the extracted knowledge through commonsense and domain knowledge, iii) the generation of explanations for human experts, iv) the ability to map human feedback into the validated representation from i), and v) the injection of some of this knowledge in a non-trained network to enable knowledge-informed representation learning. The holistic combination of causality, expressive logical inference, and representation learning, would result in a seamless integration of (neural) learning and (cognitive) reasoning that makes it possible to retain access to the inherently explainable symbolic representation without losing the power of the deep representation. The involvement of human experts in the design, validation and knowledge injection process is crucial, as the conceptual approach paves the way for a new human–ai paradigm where the human role goes beyond that of labeling data, towards the validation of neural-cognitive knowledge and processes. Show more

Keywords: Explainable AI, neuro-symbolic cycle, graph analysis, rule extraction, human-centric AI

DOI: 10.3233/NAI-240740

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-13, 2024

Authors: Sun, Ron

Article Type: Research Article

Abstract: This position paper discusses relationships among hybrid neural-symbolic models, dual-process theories, and cognitive architectures. It provides some historical backgrounds and argues that dual-process (implicit versus explicit) theories have significant implications for developing neural-symbolic (neurosymbolic) models. Furthermore, computational cognitive architectures can help to disentangle issues concerning dual-process theories and thus help the development of neural-symbolic models (in this way as well as in other ways).

Keywords: Neural-symbolic, neurosymbolic, dual process, cognitive architecture, Clarion, implicit, explicit, psychology

DOI: 10.3233/NAI-240720

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-9, 2024

Authors: Khan, M. Jaleed | Ilievski, Filip | Breslin, John G. | Curry, Edward

Article Type: Research Article

Abstract: Combining deep learning and common sense knowledge via neurosymbolic integration is essential for semantically rich scene representation and intuitive visual reasoning. This survey paper delves into data- and knowledge-driven scene representation and visual reasoning approaches based on deep learning, common sense knowledge and neurosymbolic integration. It explores how scene graph generation, a process that detects and analyses objects, visual relationships and attributes in scenes, serves as a symbolic scene representation. This representation forms the basis for higher-level visual reasoning tasks such as visual question answering, image captioning, image retrieval, image generation, and multimodal event processing. Infusing common sense knowledge, particularly …through the use of heterogeneous knowledge graphs, improves the accuracy, expressiveness and reasoning ability of the representation and allows for intuitive downstream reasoning. Neurosymbolic integration in these approaches ranges from loose to tight coupling of neural and symbolic components. The paper reviews and categorises the state-of-the-art knowledge-based neurosymbolic approaches for scene representation based on the types of deep learning architecture, common sense knowledge source and neurosymbolic integration used. The paper also discusses the visual reasoning tasks, datasets, evaluation metrics, key challenges and future directions, providing a comprehensive review of this research area and motivating further research into knowledge-enhanced and data-driven neurosymbolic scene representation and visual reasoning. Show more

Keywords: Scene graph, image representation, deep learning, common sense knowledge, neurosymbolic integration, visual reasoning

DOI: 10.3233/NAI-240719

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-24, 2024

Authors: Hitzler, Pascal | Rayan, Rushrukh | Zalewski, Joseph | Saki Norouzi, Sanaz | Eberhart, Aaron | Vasserman, Eugene Y.

Article Type: Research Article

Abstract: Deep Deductive Reasoning refers to the training and then executing of deep learning systems to perform deductive reasoning in the sense of formal, mathematical logic. We discuss why this is an interesting and relevant problem to study, and explore how hard it is as a deep learning problem. In particular, we present some of the progress made on this topic in recent years, understand some of the theoretical limitations that can be assessed from existing literature, and discuss negative results we have obtained regarding improving on the state of the art.

Keywords: Deductive reasoning, deep learning, theoretical limitations of deep learning, memory networks, transformers

DOI: 10.3233/NAI-240669

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-13, 2024

Authors: Confalonieri, Roberto | Guizzardi, Giancarlo

Article Type: Research Article

Abstract: There has been a renewed interest in symbolic AI in recent years. Symbolic AI is indeed one of the key enabling technologies for the development of neuro-symbolic AI systems, as it can mitigate the limited capabilities of black box deep learning models to perform reasoning and provide support for explanations. This paper discusses the different roles that explicit knowledge, in particular ontologies, can play in drawing intelligible explanations in neuro-symbolic AI. We consider three main perspectives in which ontologies can contribute significantly, namely reference modelling, common-sense reasoning, and knowledge refinement and complexity management. We overview some of the existing approaches …in the literature, and we position them according to these three proposed perspectives. The paper concludes by discussing some open challenges related to the adoption of ontologies in explanations. Show more

Keywords: Neuro-symbolic AI, explanations, applied ontologies

DOI: 10.3233/NAI-240754

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-15, 2024

Authors: Sha, Jingyuan | Shindo, Hikaru | Kersting, Kristian | Dhami, Devendra Singh

Article Type: Research Article

Abstract: The predicates used for Inductive Logic Programming (ILP) systems are usually elusive and need to be hand-crafted in advance, which limits the generalization of the system when learning new rules without sufficient background knowledge. Predicate Invention (PI) for ILP is the problem of discovering new concepts that describe hidden relationships in the domain. PI can mitigate the generalization problem for ILP by inferring new concepts, giving the system a better vocabulary to compose logic rules. Although there are several PI approaches for symbolic ILP systems, PI for Neuro-Symbolic-ILP (NeSy-ILP) systems that can handle 3D visual inputs to learn logical rules …using differentiable reasoning is still unaddressed. To this end, we propose a neuro-symbolic approach, NeSy-π , to invent predicates from visual scenes for NeSy-ILP systems based on clustering and extension of relational concepts, where π denotes the abbrivation of P redicate I nvention. NeSy-π processes visual scenes as input using deep neural networks for the visual perception and invents new concepts that support the task of classifying complex visual scenes. The invented concepts can be used by any NeSy-ILP system instead of hand-crafted background knowledge. Our experiments show that the NeSy-π is capable of inventing high-level concepts and solving complex visual logic patterns efficiently and accurately in the absence of explicit background knowledge. Moreover, the invented concepts are explainable and interpretable, while also providing competitive results with state-of-the-art NeSy-ILP systems. (github: https://github.com/ml-research/NeSy-PI ) Show more

Keywords: Predicate Invention, Inductive Logic Programming, Neuro-Symbolic Artificial Intelligence

DOI: 10.3233/NAI-240712

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-26, 2024

Authors: Alam, Mehwish | van Harmelen, Frank | Acosta, Maribel

Article Type: Research Article

Abstract: Embedding based Knowledge Graph (KG) completion has gained much attention over the past few years. Most of the current algorithms consider a KG as a multidirectional labeled graph and lack the ability to capture the semantics underlying the schematic information. This position paper revises the state of the art and discusses several variations of the existing algorithms for KG completion, which are discussed progressively based on the level of expressivity of the semantics utilized. The paper begins with analysing various KG completion algorithms considering only factual information such as transductive and inductive link prediction and entity type prediction algorithms. It …then revises the algorithms utilizing Large Language Models as background knowledge. Afterwards, it discusses the algorithms progressively utilizing semantic information such as class hierarchy information within the KGs and semantics represented in different description logic axioms. The paper concludes with a critical reflection on the current state of work in the community, where we argue that the aspects of semantics, rigorous evaluation protocols, and bias against external sources have not been sufficiently addressed in the literature, which hampers a more thorough understanding of advantages and limitations of existing approaches. Lastly, we provide recommendations for future directions. Show more

Keywords: Knowledge graph embeddings, semantics in knowledge graph embeddings, large language models

DOI: 10.3233/NAI-240731

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-17, 2024

Authors: Hersche, Michael | Terzić, Aleksandar | Karunaratne, Geethan | Langenegger, Jovin | Pouget, Angéline | Cherubini, Giovanni | Benini, Luca | Sebastian, Abu | Rahimi, Abbas

Article Type: Research Article

Abstract: Distributed sparse block codes (SBCs) exhibit compact representations for encoding and manipulating symbolic data structures using fixed-width vectors. One major challenge however is to disentangle, or factorize, the distributed representation of data structures into their constituent elements without having to search through all possible combinations. This factorization becomes more challenging when SBCs vectors are noisy due to perceptual uncertainty and approximations made by modern neural networks to generate the query SBCs vectors. To address these challenges, we first propose a fast and highly accurate method for factorizing a more flexible and hence generalized form of SBCs, dubbed GSBCs. Our iterative …factorizer introduces a threshold-based nonlinear activation, conditional random sampling, and an ℓ ∞ -based similarity metric. Its random sampling mechanism, in combination with the search in superposition, allows us to analytically determine the expected number of decoding iterations, which matches the empirical observations up to the GSBC’s bundling capacity. Secondly, the proposed factorizer maintains a high accuracy when queried by noisy product vectors generated using deep convolutional neural networks (CNNs). This facilitates its application in replacing the large fully connected layer (FCL) in CNNs, whereby C trainable class vectors, or attribute combinations, can be implicitly represented by our factorizer having F -factor codebooks, each with C F fixed codevectors. We provide a methodology to flexibly integrate our factorizer in the classification layer of CNNs with a novel loss function. With this integration, the convolutional layers can generate a noisy product vector that our factorizer can still decode, whereby the decoded factors can have different interpretations based on downstream tasks. We demonstrate the feasibility of our method on four deep CNN architectures over CIFAR-100, ImageNet-1K, and RAVEN datasets. In all use cases, the number of parameters and operations are notably reduced compared to the FCL. Show more

Keywords: Vector-symbolic architectures, iterative factorizer, sparse block codes, convolutional neural networks, deep learning

DOI: 10.3233/NAI-240713

Citation: Neurosymbolic Artificial Intelligence, vol. Pre-press, no. Pre-press, pp. 1-22, 2024