Taste perceptual circuit for the generation of autonomous responses in virtual creatures

BraLearning is one of the most complex cognitive processes that living beings have. This process is distributed within the brain and each independent cognitive function contributes to a small or large extent to it. Particularly, affective learning is a process that allows us to identify stimuli in the environment and discriminate them based on our tastes or preferences. Within this work we will address a type of affective evaluation generated by the perceptual circuit of taste that allows us to determine flavors from the environment and how this information becomes part of the perceptual affective evaluation. We will start this work by identifying other computational works that address affective processes oriented to taste or that involve some other affective evaluation. Subsequently, we analyze of neuroscientific information that allows us to build an informal model of the perceptual circuit of taste. Finally, we propose a cognitive architecture for the generation of autonomous responses oriented to taste in virtual creatures and we will demonstrate its functionality through an implementation of this model in some case studies through distributed modules and discuss the results obtained.     

On the Algebraizability of Annotated Logics

Annotated logics were introduced by V.S. Subrahmanian as logical foundations for computer programming. One of the difficulties of these systems from the logical point of view is that they are not structural, i.e., their consequence relations are not closed under substitutions. In this paper we give systems of annotated logics that are equivalent to those of Subrahmanian in the sense that everything provable in one type of system has a translation that is provable in the other. Moreover these new systems are structural. We prove that these systems are weakly congruential, namely, they have an infinite system of congruence 1-formulas. Moreover, we prove that an annotated logic is algebraizable (i.e., it has a finite system of congruence formulas,) if and only if the lattice of annotation constants is finite.     

Cognitive technology to capture deep computational concepts with combinators

Computational activity is now recognized as a natural science, and computational and information processes have been discovered in the deep structures of many areas. Computations in the natural world were present long before the invention of computers, but a remarkable shift in understanding its fundamental nature occurs, in fact, before our eyes. The present moment, in fact, is a transition from the concept of computer science as an artificial science to the understanding that information processes are abundant in nature. Computing is recognized as a natural science that studies natural and artificial information processes.In everyday computing, operations are performed on the individual generators, with little attention paid to their internal structure. However, many common operations consist of more primitive constructions connected by a combination mode. The interaction of information processes and corresponding structures is carried out in an environment of “applicative interaction”, their applications to each other, and the study of the properties of this environment allows us to understand the nature of the computations.In the present work, the main attention is paid to elucidating the technological features of computations with individual generators, or objects. Their interaction is considered in an applicative environment, which allows us to elucidate the internal structure of ordinary operations, the knowledge of which allows us to understand their properties. The choice of initial constant generators, considered as generic ones and expressed by combinators, is discussed. These initial generators are used as the main “building blocks” that occur within the larger blocks of the applicative environment in interaction with each other. As a result of the interaction, constructions arise that give representative sets of ordinary operators and embedded computing systems.     

Logic Without Contraction as Based on Inclusion and Unrestricted Abstraction

On the one hand, the absence of contraction is a safeguard against the logical (property theoretic) paradoxes; but on the other hand, it also disables inductive and recursive definitions, in its most basic form the definition of the series of natural numbers, for instance. The reason for this is simply that the effectiveness of a recursion clause depends on its being available after application, something that is usually assured by contraction. This paper presents a way of overcoming this problem within the framework of a logic based on inclusion and unrestricted abstraction, without any form of extensionality. This revised version was published online in June 2006 with corrections to the Cover Date.     

Abductive reasoning in neural-symbolic systems

Abduction is or subsumes a process of inference. It entertains possible hypotheses and it chooses hypotheses for further scrutiny. There is a large literature on various aspects of non-symbolic, subconscious abduction. There is also a very active research community working on the symbolic (logical) characterisation of abduction, which typically treats it as a form of hypothetico-deductive reasoning. In this paper we start to bridge the gap between the symbolic and sub-symbolic approaches to abduction. We are interested in benefiting from developments made by each community. In particular, we are interested in the ability of non-symbolic systems (neural networks) to learn from experience using efficient algorithms and to perform massively parallel computations of alternative abductive explanations. At the same time, we would like to benefit from the rigour and semantic clarity of symbolic logic. We present two approaches to dealing with abduction in neural networks. One of them uses Connectionist Modal Logic and a translation of Horn clauses into modal clauses to come up with a neural network ensemble that computes abductive explanations in a top-down fashion. The other combines neural-symbolic systems and abductive logic programming and proposes a neural architecture which performs a more systematic, bottom-up computation of alternative abductive explanations. Both approaches employ standard neural network architectures which are already known to be highly effective in practical learning applications. Differently from previous work in the area, our aim is to promote the integration of reasoning and learning in a way that the neural network provides the machinery for cognitive computation, inductive learning and hypothetical reasoning, while logic provides the rigour and explanation capability to the systems, facilitating the interaction with the outside world. Although it is left as future work to determine whether the structure of one of the proposed approaches is more amenable to learning than the other, we hope to have contributed to the development of the area by approaching it from the perspective of symbolic and sub-symbolic integration.

Paraconsistent Informational Logic

We introduce a Paraconsistent Informational Logic that formalizes the idea of conjectures which are acceptable as to the quality and the variety of the information that they convey with respect to a given theory T, even if they are classically inconsistent with T. The work constitutes an extension of a previously developed Informational Logic for classical frameworks, where a new notion of logical entropy measure H on formulas and on proofs plays a central role.     

My beliefs about your beliefs: a case study in theory of mind and epistemic logic

We model three examples of beliefs that agents may have about other agents’ beliefs, and provide motivation for this conceptualization from the theory of mind literature. We assume a modal logical framework for modelling degrees of belief by partially ordered preference relations. In this setting, we describe that agents believe that other agents do not distinguish among their beliefs (‘no preferences’), that agents believe that the beliefs of other agents are in part as their own (‘my preferences’), and the special case that agents believe that the beliefs of other agents are exactly as their own (‘preference refinement’). This multi-agent belief interaction is frame characterizable. We provide examples for introspective agents. We investigate which of these forms of belief interaction are preserved under three common forms of belief revision.     

Introspective forgetting

We model the forgetting of propositional variables in a modal logical context where agents become ignorant and are aware of each others’ or their own resulting ignorance. The resulting logic is sound and complete. It can be compared to variable-forgetting as abstraction from information, wherein agents become unaware of certain variables: by employing elementary results for bisimulation, it follows that beliefs not involving the forgotten atom(s) remain true. The work for this publication was mainly carried out while Hans van Ditmarsch was associated to: Institut de Recherche en Informatique, Université Paul Sabatier, France.     

Heuristic thinking and human intelligence: a commentary on Marewski,Gaissmaier and Gigerenzer

Marewski, Gaissmaier and Gigerenzer (2009) present a review of research on fast and frugal heuristics, arguing that complex problems are best solved by simple heuristics, rather than the application of knowledge and logical reasoning. We argue that the case for such heuristics is overrated. First, we point out that heuristics can often lead to biases as well as effective responding. Second, we show that the application of logical reasoning can be both necessary and relatively simple. Finally, we argue that the evidence for a logical reasoning system that co-exists with simpler heuristic forms of thinking is overwhelming. Not only is it implausible a priori that we would have evolved such a system that is of no use to us, but extensive evidence from the literature on dual processing in reasoning and judgement shows that many problems can only be solved when this form of reasoning is used to inhibit and override heuristic thinking.     

How Do Logics Explain?

Anti-exceptionalists about logic maintain that it is continuous with the empirical sciences. Taking anti-exceptionalism for granted, we argue that traditional approaches to explanation are inadequate in the case of logic. We argue that Andrea Woody's functional analysis of explanation is a better fit with logical practice and accounts better for the explanatory role of logical theories.     

Symbolism and literalism in anthropology

Conclusion We have considered two strategies for using native utterances as evidence for assigning native beliefs. We have shown that each of these two strategies (literalism and symbolism) can avoid the logical difficulties mentioned in section 1 — so long, at least, as we employ an account of the logical form of belief sentences developed by Burdick. We have also considered the methodological principles which provide the basis for translational practice. Based on our consideration of these principles, we then argued that we must prefer the literalist strategy for attributing beliefs. Only the literalist strategy enables us to provide a recursive account of the significance of native utterances, and only the literalist strategy enables us to maximize the truth of our claims about native beliefs.This paper was written mainly by D'Agostino, who supplied the background in philosophy of anthropology, with the assistance of Burdick, chiefly on matters relating to philosophy of language. We are extremely grateful to Susan Haack for substantial help with significant improvements. We also wish to thank J. J. C. Smart and Stanley Benn for comments.     

Reasoning about update logic

Logical frameworks for analysing the dynamics of information processing abound [4, 5, 8, 10, 12, 14, 20, 22]. Some of these frameworks focus on the dynamics of the interpretation process, some on the dynamics of the process of drawing inferences, and some do both of these. Formalisms galore, so it is felt that some conceptual streamlining would pay off.This paper is part of a larger scale enterprise to pursue the obvious parallel between information processing and imperative programming. We demonstrate that logical tools from theoretical computer science are relevant for the logic of information flow. More specifically, we show that the perspective of Hoare logic [13, 18] can fruitfully be applied to the conceptual simplification of information flow logics.     

The Russian Cards Problem

Suppose we have a stack of cards that is divided over some players. For certain distributions of cards it is possible to communicate your hand of cards to another player by public announcements, without yet another player learning any of your cards. A solution to this problem consists of some sequence of announcements and is called an exchange. It is called a direct exchange if it consists of (the minimum of) two announcements only. The announcements in an exchange have a special form: they are safe communications, an interesting new form of update. Certain unsafe communications turn out to be unsuccessful updates. A communication is a public announcement that is known to be true. Each communication may be about a set of alternative card deals only, and even about a set of alternatives to the communicating player's own hand only. We list the direct exchanges for a deal of seven cards where the two players holding three cards communicate their hands to each other. Our work may be applicable to the design of cryptographic protocols.     

The logic of the medical research article

As do all forms of science, medical theories have a factual as well as a logical basis. New information is presented in medical research articles. These papers have three separate arguments: the argument of the hypothesis, the argument of the experimental protocol, and the argument of the hypothesis's judgment. These arguments may be examples of the hypothetico-deductive or confirmational model of scientific inference. The logical form of these arguments are informal and inductive rather than formal and deductive. Understanding the nature of the logic of the medical research article may help avoid erroneous conclusions.     

A new experimental method to identify the process of logical reasoning

Abstract: A new method to identify the process of logical reasoning is presented. In spite of its indispensability and importance, we have had few methods to identify a subject's reasoning process, except that of using verbal protocol data. In this paper, for the purpose of objective identification of the reasoning process, we propose a new method to obtain the subject's reasoning process, in terms of a resolution tree for a task of which the logical structure can be written by first-order predicate logic. The results of an experiment using this method are presented. They revealed some interesting features of human reasoning such as, large differences between subjects, remarkable parallel processes, and the existence of subgoals for each subject.     

The enduring scandal of deduction

Deductive inference is usually regarded as being “tautological” or “analytical”: the information conveyed by the conclusion is contained in the information conveyed by the premises. This idea, however, clashes with the undecidability of first-order logic and with the (likely) intractability of Boolean logic. In this article, we address the problem both from the semantic and the proof-theoretical point of view. We propose a hierarchy of propositional logics that are all tractable (i.e. decidable in polynomial time), although by means of growing computational resources, and converge towards classical propositional logic. The underlying claim is that this hierarchy can be used to represent increasing levels of “depth” or “informativeness” of Boolean reasoning. Special attention is paid to the most basic logic in this hierarchy, the pure “intelim logic”, which satisfies all the requirements of a natural deduction system (allowing both introduction and elimination rules for each logical operator) while admitting of a feasible (quadratic) decision procedure. We argue that this logic is “analytic” in a particularly strict sense, in that it rules out any use of “virtual information”, which is chiefly responsible for the combinatorial explosion of standard classical systems. As a result, analyticity and tractability are reconciled and growing degrees of computational complexity are associated with the depth at which the use of virtual information is allowed.     

Using logical relevance for question answering

We propose a novel method of determining the appropriateness of an answer to a question through a proof of logical relevance rather than a logical proof of truth. We define logical relevance as the idea that answers should not be considered as absolutely true or false in relation to a question, but should be considered true more flexibly in a sliding scale of aptness. This enables us to reason rigorously about the appropriateness of an answer even in cases where the sources we are getting answers from are incomplete or inconsistent or contain errors. We show how logical relevance can be implemented through the use of measured simplification, a form of constraint relaxation, in order to seek a logical proof than an answer is in fact an answer to a particular question. We then give an example of such an implementation providing a set of specific rules for this purpose.     

Minimal Requirements for the Emergence of Learned Signaling

The emergence of signaling systems has been observed in numerous experimental and real‐world contexts, but there is no consensus on which (if any) shared mechanisms underlie such phenomena. A number of explanatory mechanisms have been proposed within several disciplines, all of which have been instantiated as credible working models. However, they are usually framed as being mutually incompatible. Using an exemplar‐based framework, we replicate these models in a minimal configuration which allows us to directly compare them. This reveals that the development of optimal signaling is driven by similar mechanisms in each model, which leads us to propose three requirements for the emergence of conventional signaling. These are the creation and transmission of referential information, a systemic bias against ambiguity, and finally some form of information loss. Considering this, we then discuss some implications for theoretical and experimental approaches to the emergence of learned communication.