A Two-Tiered Reasoning and Learning Architecture

arXiv preprint cs/9508102, 1995

Abstract: Learning and reasoning are both aspects of what is considered to be intelligence. Their studies within AI have been separated historically, learning being the topic of machine learning and neural networks, and reasoning falling under classical (or symbolic) AI. However, learning and reasoning are in many ways interdependent. This paper discusses the nature of some of these interdependencies and proposes a general framework called FLARE, that combines inductive learning using prior knowledge together with reasoning ...

Connection Science

In this work we present a knowledge-based system equipped with a hybrid, cognitively inspired architecture for the representation of conceptual information. The proposed system aims at extending the classical representational and reasoning capabilities of the ontology-based frameworks towards the realm of the prototype theory. It is based on a hybrid knowledge base, composed of a classical symbolic component (grounded on a formal ontology) with a typicality based one (grounded on the conceptual spaces framework). The resulting system attempts to reconcile the heterogeneous approach to the concepts in Cognitive Science with the dual process theories of reasoning and rationality. The system has been experimentally assessed in a conceptual categorisation task where common sense linguistic descriptions were given in input, and the corresponding target concepts had to be identified. The results show that the proposed solution substantially extends the representational and reasoning ‘conceptual’ capabilities of standard ontology-based systems.

2004

Many classification systems rely on clustering techniques in which a collection of training examples is provided as an input, and a number of clusters c1, . . . cm modeling some concept C results as an output, such that every cluster ci is labeled as positive or negative. In such a setting clusters can overlap, and a new unlabeled instance can be assigned to more than one cluster with conflicting labels. In the literature, such a case is usually solved non-deterministically by making a random choice. This paper introduces a novel, hybrid approach to solve the above problem by combining a neural network N along with a background theory T specified in defeasible logic programming (DeLP) which models preference criteria for performing clustering.

In this paper, we point out the long lasting debate between two main approaches of Artificial Intelligence: Symbolic AI and connectionist AI. The reasoning of the connectionist AI defenders depend on the indication of necessity of neuron modelling for knowledge representation and hence ignoring symbolic AI which fails in doing that. On the other hand, symbolic AI presents better models of representing knowledge, and higher reasoning capabilities. So we will investigate the issue of whether connectionists might be justified in their arguments since symbolic AI has no intent to consider neuronal level knowledge representation and manipulations and fails most of the time when agents it considers are embodied in a real world environment where as connectionist approaches mainly consider embodiment where there are yet no serious signs of how high level reasoning might have emerged from low level functioning of interconnected neurons.

Machine Learning, 1993

This article describes a framework for the deep and dynamic integration of learning strategies. The framework is based on the idea that each single-strategy learning method is ultimately the result of certain elementary inferences (like deduction, analogy, abduction, generalization, specialization, abstraction, concretion, etc.). Consequently, instead of integrating learning strategies at a macro level, we propose to integrate the different inference types that generate individual learning strategies. The article presents a concept-learning and theory-revision method that was developed in this framework. It allows the system to learn from one or from several (positive and/or negative) examples, and to both generalize and specialize its knowledge base. The method integrates deeply and dynamically different learning strategies, depending on the relationship between the input information and the knowledge base. It also behaves as a single-strategy learning method whenever the applicability conditions of such a method are satisfied.

Lecture Notes in Computer Science, 2013

This paper reports the preliminary experiments in a generalpurpose reasoning system to carry out natural language comprehension and production using a reasoning-learning mechanism designed to realize general intelligence.

2009

This research is an exploration of the phenomenon of "intuition" in the context of artificial intelligence (AI). In this work, intuition was considered as the human capacity to make decisions under situations in which the available knowledge was usually low in quality: inconsistent and of varying levels of certainty. The objectives of this study were to characterize some of the aspects of human intuitive thought and to model these aspects in a computational approach.

2016

Several investigations have been developed around analogies based reasoning in different domains, however the analogy between arguments has not been deeply explored. A semiformal way to express these patterns of reasoning were proposed by Walton, through argument schemes from analogy. From this, it is possible to propose computable approximations for comparing arguments. In this paper we introduce a formalism based on the comparison of arguments through descriptors or labels which describes an aspect that the argument refers to. This formalism allows us classifying similar arguments considering the natural descriptors of them, in a specific context.

English version of my French "Reseaux de neurones capables de raisonner", Dossier Pour la Science (special issue of the French edition of the Scientific American), October/December 2005, 97-101., 2005

This paper gives a very short (and accessible) survey of how classical and nonmonotonic logic relate to neural networks in general, and on how neural networks might be able to carry out logical reasoning in particular.

2017

This paper presents an ongoing research project called “natural logic” and makes the case that it is relevant to AI, Computational Linguistics, and Cognitive Science. We propose to add some of the natural logic modules which have already been developed to existing NLP systems. We see our approach as complementing and augmenting data-driven approaches exemplified by IBM’s Watson. We give a brief introduction to natural logic and present examples of proofs that can be given in a working system. We furthermore introduce monotonic logic, another promising approach for extracting information from sentences that contain quantifiers. We finish the paper by presenting some early work that integrates syllogistic reasoning into exsiting NLP systems. Introduction The history of logic and AI is a checkered one. Starting with huge optimism, the idea of applying logic in AI and NLP is very much a minority one today. The 2015 report of the One Hundred Year Study on Artificial Intelligence states t...

AAAI Spring Symposium Combining Machine Learning with Knowledge Engineering, 2019

In our poster, we will introduce new datasets in propositional logic and first-order logic that can be used for learning to reason, and present some initial results on systems that use this data.

2019

This paper proposes a tentative and original survey of meeting points between Knowledge Representation and Reasoning (KRR) and Machine Learning (ML), two areas which have been developing quite separately in the last three decades. Some common concerns are identified and discussed such as the types of used representation, the roles of knowledge and data, the lack or the excess of information, or the need for explanations and causal understanding. Then some methodologies combining reasoning and learning are reviewed (such as inductive logic programming, neuro-symbolic reasoning, formal concept analysis, rule-based representations and ML, uncertainty in ML, or case-based reasoning and analogical reasoning), before discussing examples of synergies between KRR and ML (including topics such as belief functions on regression, EM algorithm versus revision, the semantic description of vector representations, the combination of deep learning with high level inference, knowledge graph completi...

2000

Hybrid connectionist symbolic systems have been the subject of much recent research in AI. By focusing on the implementation of high-level human cognitive processes (eg, rule-based inference) on low-level, brain-like structures (eg, neural networks), hybrid systems inherit both the efficiency of connectionism and the comprehensibility of symbolism. This paper presents the Basic Reasoning Applicator Implemented as a Neural Network (BRAINN).

2009

Abstract An increasing amount of work in AI in general, and case-based reasoning in particular, is addressing means of integrating different types of reasoning methods for building and maintaining AI systems. This paper first takes a step back in an attempt to identify a descriptive and comparative framework for the corresponding methods. It is argued that the knowledge level is the appropriate level for describing the behaviour of an intended system, and for identifying its methods and knowledge components.

Case-based reasoning systems are systems that take advantage from experience, and thus strain after adapting to their environment. They show different ways of achieving this goal. They all share this peculiarity to be memory-driven systems. Thus, adaptability in a casebased reasoning system must be studied from the point of view of the memory. If case-based reasoners generally are devoted to the realization of a single task, the need for such systems to perform various tasks questions how to organize their memory to permit them to be taskadaptive. Moreover they must be able to adapt to several types of cognitive tasks, analysis tasks as well as synthesis tasks. For that purpose, a case-based reasoning system adaptive to the cognitive task is presented in this paper. Its adaptability comes from its memory composition, both cases and concepts, and from its hierarchical memory organization, based on multiple points of view, some of them associated to the various cognitive tasks it performs. For analytic tasks, the most specific cases are preferably used for the reasoning. For synthesis tasks, the most specific concepts, learnt by conceptual clustering, are used. So intensifying the learning inferences during case-based reasoning, and especially the synthetic learning ones, enlarges the application range of case-based reasoning to true synthesis tasks. An example of this system abilities, in the domain of eating disorders in psychiatry, is then briefly presented.