Connectionist models provide a promising alternative to the traditional computational approach that has for several decades dominated cognitive science and artificial intelligence, although the nature of connectionist models and their relation to symbol processing remains controversial. Connectionist models can be characterized by three general computational features: distinct layers of interconnected units, recursive rules for updating the strengths of the connections during learning, and “simple” homogeneous computing elements. Using just these three features one can construct surprisingly elegant and powerful models of (...) memory, perception, motor control, categorization, and reasoning. What makes the connectionist approach unique is not its variety of representational possibilities or its departure from explicit rule-based models, or even its preoccupation with the brain metaphor. Rather, it is that connectionist models can be used to explore systematically the complex interaction between learning and representation, as we try to demonstrate through the analysis of several large networks. (shrink)

The inductive category formation framework, an influential set of theories of learning in psychology and artificial intelligence, is deeply flawed. In this framework a set of necessary and sufficient features is taken to define a category. Such definitions are not functionally justified, are not used by people, and are not inducible by a learning system. Inductive theories depend on having access to all and only relevant features, which is not only impossible but begs a key question in learning. The crucial (...) roles of other cognitive processes (such as explanation and credit assignment) are ignored or oversimplified. Learning necessarily involves pragmatic considerations that can only be handled by complex cognitive processes.We provide an alternative framework for learning according to which category definitions must be based on category function. The learning system invokes other cognitive processes to accomplish difficult tasks, makes inferences, analyses and decides among potential features, and specifies how and when categories are to be generated and modified. We also examine the methodological underpinnings of the two approaches and compare their motivations. (shrink)

Cognitive science is a child of the 1950s, the product of a time when psychology, anthropology and linguistics were redefining themselves and computer science and neuroscience as disciplines were coming into existence. Psychology could not participate in the cognitive revolution until it had freed itself from behaviorism, thus restoring cognition to scientific respectability. By then, it was becoming clear in several disciplines that the solution to some of their problems depended crucially on solving problems traditionally allocated to other disciplines. Collaboration (...) was called for: this is a personal account of how it came about. (shrink)

In this paper I start from a definition of “culture of the artificial” which might be stated by referring to the background of philosophical, methodological, pragmatical assumptions which characterizes the development of the information processing analysis of mental processes and of some trends in contemporary cognitive science: in a word, the development of AI as a candidate science of mind. The aim of this paper is to show how (with which plausibility and limitations) the discovery of the mentioned background might (...) be dated back to a period preceding the cybernetic era, the decade 1930–1940 at least. Therefore a somewhat detailed analysis of Hull's “robot approach” is given, as well as of some of its independent and future developments. -/- Reprinted in R.L. Chrisley (ed.), Artificial Intelligence: Critical Concepts in Cognitive Science, vol. 1, Routledge, London and New York, 2000, pp. 301-326. (shrink)

Explaining the mind by building machines with minds runs into the other-minds problem: How can we tell whether any body other than our own has a mind when the only way to know is by being the other body? In practice we all use some form of Turing Test: If it can do everything a body with a mind can do such that we can't tell them apart, we have no basis for doubting it has a mind. But what is (...) "everything" a body with a mind can do? Turing's original "pen-pal" version (the TT) only tested linguistic capacity, but Searle has shown that a mindless symbol-manipulator could pass the TT undetected. The Total Turing Test (TTT) calls for all of our linguistic and robotic capacities; immune to Searle's argument, it suggests how to ground a symbol manipulating system in the capacity to pick out the objects its symbols refer to. No Turing Test, however, can guarantee that a body has a mind. Worse, nothing in the explanation of its successful performance requires a model to have a mind at all. Minds are hence very different from the unobservables of physics (e.g., superstrings); and Turing Testing, though essential for machine-modeling the mind, can really only yield an explanation of the body. (shrink)

Fifty years of effort in artificial intelligence (AI) and the formalization of legal reasoning have produced both successes and failures. Considerable success in organizing and displaying evidence and its interrelationships has been accompanied by failure to achieve the original ambition of AI as applied to law: fully automated legal decision-making. The obstacles to formalizing legal reasoning have proved to be the same ones that make the formalization of commonsense reasoning so difficult, and are most evident where legal reasoning has to (...) meld with the vast web of ordinary human knowledge of the world. Underlying many of the problems is the mismatch between the discreteness of symbol manipulation and the continuous nature of imprecise natural language, of degrees of similarity and analogy, and of probabilities. (shrink)

This book reports on the results of the third edition of the premier conference in the field of philosophy of artificial intelligence, PT-AI 2017, held on November 4 - 5, 2017 at the University of Leeds, UK. It covers: advanced knowledge on key AI concepts, including complexity, computation, creativity, embodiment, representation and superintelligence; cutting-edge ethical issues, such as the AI impact on human dignity and society, responsibilities and rights of machines, as well as AI threats to humanity and AI safety; (...) and cutting-edge developments in techniques to achieve AI, including machine learning, neural networks, dynamical systems. The book also discusses important applications of AI, including big data analytics, expert systems, cognitive architectures, and robotics. It offers a timely, yet very comprehensive snapshot of what is going on in the field of AI, especially at the interfaces between philosophy, cognitive science, ethics and computing. (shrink)

The work done in the philosophy of modeling by Vaihinger (1876), Craik (1943), Rosenblueth and Wiener (1945), Apostel (1960), Minsky (1965), Klaus (1966) and Stachowiak (1973) is still almost completely neglected in the mainstream literature. However, this work seems to contain original ideas worth to be discussed. For example, the idea that diverse functions of models can be better structured as follows: in fact, models perform only a single function – they are replacing their target systems, but for different purposes. (...) Another example: the idea that all of cognition is cognition in models or by means of models. Even perception, reflexes and instincts (animal and human) can be best analyzed as modeling. The paper presents an analysis of the above-mentioned work. (shrink)

In attempts to formulate a computational understanding of brain function, one of the fundamental concerns is the data structure by which the brain represents information. For many decades, a conceptual framework has dominated the thinking of both brain modelers and neurobiologists. That framework is referred to here as "classical neural networks." It is well supported by experimental data, although it may be incomplete. A characterization of this framework will be offered in the next section. Difficulties in modeling important functional aspects (...) of the brain on the basis of classical neural networks alone have led to the recognition that another, general mechanism must be invoked to explain brain function. That mechanism I call " binding." Binding by neural signal synchrony had been mentioned several times in the liter ature before it was fully formulated as a general phenomenon. Although experimental evidence for neural syn chrony was soon found, the idea was largely ignored for many years. Only recently has it become a topic of animated discussion. In what follows, I will summarize the nature and the roots of the idea of binding, especially of temporal binding, and will discuss some of the objec tions raised against it. (shrink)

I propose that the successes and contributions of reinforcement learning urge us to see the mind in a new light, namely, to recognise that the mind is fundamentally evaluative in nature.

Effective conditioning requires a correlation between the experimenter's definition of a response and an organism's, but an animal's perception of its behavior differs from ours. These experiments explore various definitions of the response, using the slopes of learning curves to infer which comes closest to the organism's definition. The resulting exponentially weighted moving average provides a model of memory that is used to ground a quantitative theory of reinforcement. The theory assumes that: incentives excite behavior and focus the excitement on (...) responses that are contemporaneous in memory. The correlation between the organism's memory and the behavior measured by the experimenter is given by coupling coefficients, which are derived for various schedules of reinforcement. The coupling coefficients for simple schedules may be concatenated to predict the effects of complex schedules. The coefficients are inserted into a generic model of arousal and temporal constraint to predict response rates under any scheduling arrangement. The theory posits a response-indexed decay of memory, not a time-indexed one. It requires that incentives displace memory for the responses that occur before them, and may truncate the representation of the response that brings them about. As a contiguity-weighted correlation model, it bridges opposing views of the reinforcement process. By placing the short-term memory of behavior in so central a role, it provides a behavioral account of a key cognitive process. (shrink)

We propose a non-representationalist framework for deep learning relying on a novel method computational phenomenology, a dialogue between the first-person perspective (relying on phenomenology) and the mechanisms of computational models. We thereby propose an alternative to the modern cognitivist interpretation of deep learning, according to which artificial neural networks encode representations of external entities. This interpretation mainly relies on neuro-representationalism, a position that combines a strong ontological commitment towards scientific theoretical entities and the idea that the brain operates on symbolic (...) representations of these entities. We proceed as follows: after offering a review of cognitivism and neuro-representationalism in the field of deep learning, we first elaborate a phenomenological critique of these positions; we then sketch out computational phenomenology and distinguish it from existing alternatives; finally we apply this new method to deep learning models trained on specific tasks, in order to formulate a conceptual framework of deep-learning, that allows one to think of artificial neural networks’ mechanisms in terms of lived experience. (shrink)