This is the guest editor's introduction to a JETAI special issue on philosophical foundations of AI.

Epistemologists have debated at length whether scientific discovery is a rational and logical process. If it is, according to the Artificial Intelligence hypothesis, it should be possible to write computer programs able to discover laws or theories; and if such programs were written, this would definitely prove the existence of a logic of discovery. Attempts in this direction, however, have been unsuccessful: the programs written by Simon's group, indeed, infer famous laws of physics and chemistry; but having found no new (...) law, they cannot properly be considered discovery machines. The programs written in the Turing tradition, instead, produced new and useful empirical generalization, but no theoretical discovery, thus failing to prove the logical character of the most significant kind of discoveries. A new cognitivist and connectionist approach by Holland, Holyoak, Nisbett and Thagard, looks more promising. Reflection on their proposals helps to understand the complex character of discovery processes, the abandonment of belief in the logic of discovery by logical positivists, and the necessity of a realist interpretation of scientific research. (shrink)

In CyberPhilosophy: The Intersection of Philosophy and Computing, edited by James H. Moor and Terrell Ward Bynum (Oxford, UK: Blackwell, 2002), 66-77. Also in Metaphilosophy 33.1/2 (2002): 70-82.

The problem of valid induction could be stated as follows: are we justified in accepting a given hypothesis on the basis of observations that frequently confirm it? The present paper argues that this question is relevant for the understanding of Machine Learning, but insufficient. Recent research in inductive reasoning has prompted another, more fundamental question: there is not just one given rule to be tested, there are a large number of possible rules, and many of these are somehow confirmed by (...) the data — how are we to restrict the space of inductive hypotheses and choose effectively some rules that will probably perform well on future examples? We analyze if and how this problem is approached in standard accounts of induction and show the difficulties that are present. Finally, we suggest that the explanation-based learning approach and related methods of knowledge intensive induction could be, if not a solution, at least a tool for solving some of these problems. (shrink)

_This paper outlines the case for believing that we will have superhuman artificial intelligence_ _within the first third of the next century. It looks at different estimates of the processing power of_ _the human brain; how long it will take until computer hardware achieve a similar performance;_ _ways of creating the software through bottom-up approaches like the one used by biological_ _brains; how difficult it will be for neuroscience figure out enough about how brains work to_ _make this approach work; (...) and how fast we can expect superintelligence to be developed once_ _there is human-level artificial intelligence._. (shrink)

Researchers in artificial intelligence (AI) Moore’s law states that computational complexity of the models is still far below that and artificial life (Alife) are interested resources for a fixed price roughly double of any living system. New experiments in evo- in understanding the properties of liv- every 18 months. From about 1975 into the lution simulate spatially isolated populations ing organisms so that they can build artificial early 1990s all the gains of Moore’s law went to investigate speciation. Over the (...) past few systems that exhibit these properties for into the changeover from the centralized years, new directions have emerged in AI⁵, in useful purposes. AI researchers are interest- mainframe to the individual computer on attempts to implement artificial creatures in ed mostly in perception, cognition and your desk, accommodating a vastly simulated or physical environments. generation of action (Box 1), whereas Alife increased number of users. The amount of Often called the behaviour-based focuses on evolution, reproduction, computing power available to the individual approach, this new mode of thought involves morphogenesis and metabolism (Box 2). scientist did not change that much, although the connection of perception to action with Neither of these disciplines is a conventional the price came down by a factor of a little in the way of intervening representa- science; rather, they are a mixture of science thousand. But since the early 1990s, all of tional systems. Rather than relying on and engineering. Despite, or perhaps Moore’s law has gone into increasing the per- search, this approach relies on the correct because of, this hybrid structure, both disci- formance of the workstation itself. short, fast connections being present plines have been very successful and our And both AI and Alife have benefited from between sensory and motor modules. world is full of their products. this shift. Behaviour-based approaches began with Every time we use a computer we use Increased computer power has enabled insect models, but more recently they have algorithms and techniques developed by AI search-based AI to push ahead with been extended to humanoid robots⁶ — researchers. (shrink)

wide variety of things. It covers the capacity to carry out deductive inferences, to make.

Applied Artificial Intelligence, 21, 2007, pp. 259-279.

Lecture Notes in Artificial Intelligence, vol. 4155, Springer, Berlin-Heidelberg, 2006, pp. 1-23. PDF.

Actual AI research began auspiciously around 1955 with Allen Newell and Herbert Simon's work at the RAND Corporation. Newell and Simon proved that computers could do more than calculate. They demonstrated that computers were physical symbol systems whose symbols could be made to stand for anything, including features of the real world, and whose programs could be used as rules for relating these features. In this way computers could be used to simulate certain important aspects intelligence. Thus the information-processing model (...) of the mind was born. But, looking back over these fifty years, it seems that theoretical AI with its promise of a robot like HAL appears to be a perfect example of what Imre Lakatos has called a "degenerating research program". (shrink)

The extent to which concepts, memory, and planning are necessary to the simulation of intelligent behavior is a fundamental philosophical issue in Artificial Intelligence. An active and productive segement of the AI community has taken the position that multiple low-level agents, properly organized, can account for high-level behavior. Empirical research on these questions with fully operational systems has been restricted to mobile robots that do simple tasks. This paper recounts experiments with Hoyle, a system in a cerebral, rather than a (...) physical, domain. The program learns to perform well and quickly, often outpacing its human creators at two-person, perfect information board games. Hoyle demonstrates that a surprising amount of intelligent behavior can be treated as if it were situation-determined, that often planning is unnecessary, and that the memory required to support this learning is minimal. Concepts, however, are crucial to this reactive program's ability to learn and perform. (shrink)

This paper supports the view that the ongoing shift from orthodox to embodied-embedded cognitive science has been significantly influenced by the experimental results generated by AI research. Recently, there has also been a noticeable shift toward enactivism, a paradigm which radicalizes the embodied-embedded approach by placing autonomous agency and lived subjectivity at the heart of cognitive science. Some first steps toward a clarification of the relationship of AI to this further shift are outlined. It is concluded that the success of (...) enactivism in establishing itself as a mainstream cognitive science research program will depend less on progress made in AI research and more on the development of a phenomenological pragmatics. (shrink)

Self-improvement was one of the aspects of AI proposed for study in the 1956 Dartmouth conference. Turing proposed a “child machine” which could be taught in the human manner to attain adult human-level intelligence. In latter days, the contention that an AI system could be built to learn and improve itself indefinitely has acquired the label of the bootstrap fallacy. Attempts in AI to implement such a system have met with consistent failure for half a century. Technological optimists, however, have (...) maintained that a such system is possible, producing, if implemented, a feedback loop that would lead to a rapid exponential increase in intelligence. We examine the arguments for both positions and draw some conclusions. (shrink)

Contributors: Rodney A. Brooks, Paul M. Churchland, Andy Clark, Daniel C. Dennett, Hubert L. Dreyfus, Jerry A. Fodor, Joseph Garon, John Haugeland, Marvin...

The idea that human thinking and machine computing are "radically the same" provides the central theme for this marvelously lucid and witty book on...

This is a volume of original essays written by philosophers and scientists and dealing with philosophical questions arising from work in evolutionary biology and artificial intelligence. In recent years both of these areas have been the focus for attempts to provide a scientific, model of a wide range of human capacities - most prominently perhaps in sociobiology and cognitive psychology. The book therefore examines a number of issues related to the search for a 'naturalistic' or scientific account of human experience (...) and behaviour. Some of the essays deal with the application of such models to particular behaviour, stressing the problems raised by consciousness, and the information to be derived from the differing capacities of animals and people; others consider more general questions about the logic of the explanations provided by these kinds of approach. The volume continues the informal series stemming from meetings sponsored by the Thyssen Foundation. (shrink)

The “grand problem” of AI has always been to build artificial agents of human-level intelligence, capable of operating in environments of real-world complexity. OSCAR is a cognitive architecture for such agents, implemented in LISP. OSCAR is based on my extensive work in philosophy concerning both epistemology and rational decision making. This paper provides a detailed overview of OSCAR. The main conclusions are that such agents must be capablew of operating against a background of pervasive ignorance, because the real world is (...) too complex for them to know more than a small fraction of what is true. This is handled by giving the agent the power to reason defeasibily. The OSCAR system of defeasible reasoning is sketched. It is argued that if epistemic cognition must be defeasible, planning must also be done defeasibly, and the best way to do that is to reason defeasibly about plans. A sketch is given about how this might work. (shrink)

Stuart Russell [14] describes rational agents as --œthose that do the right thing--�. The problem of designing a rational agent then becomes the problem of figuring out what the right thing is. There are two approaches to the latter problem, depending upon the kind of agent we want to build. On the one hand, anthropomorphic agents are those that can help human beings rather directly in their intellectual endeavors. These endeavors consist of decision making and data processing. An agent that (...) can help humans in these enterprises must make decisions and draw conclusions that are rational by human standards of rationality. Anthropomorphic agents can be contrasted with goal-oriented agents --” those that can carry out certain narrowly-defined tasks in the world. Here the objective is to get the job done, and it makes little difference how the agent achieves its design goal. (shrink)

Intuitive conceptions guide practice, but practice reciprocally reshapes intuition. The intuitive conception of intelligence in AI was originally highly anthropocentric. However, the internal dynamics of AI research have resulted in a divergence from anthropocentric concerns. In particular, the increasing emphasis on commonsense knowledge and peripheral intelligence (perception and movement) in effect constitutes an incipient reorientation of intuitions about the nature of intelligence in a non-anthropocentric direction. I argue that this conceptual shift undermines Joseph Weizenbaum's claim that the project of artificial (...) intelligence is inherently dehumanizing. (shrink)

1 Matching Templates and Feature Analysers. 2 Modes of Perception in Left and Right Cerebral Hemispheres. 3 Identification and Recognition. 4 Chess Plying Machines.

The peculiarity of the relationship between philosophy and Artificial Intelligence (AI) has been evidenced since the advent of AI. This paper aims to put the basis of an extended and well founded philosophy of AI: it delineates a multi-layered general framework to which different contributions in the field may be traced back. The core point is to underline how in the same scenario both the role of philosophy on AI and role of AI on philosophy must be considered. Moreover, this (...) framework is revised and extended in the light of the consideration of a type of multiagent system devoted to afford the issue of scientific discovery both from a conceptual and from a practical point of view. (shrink)

Since 1991 the author has been Professor of Artificial Intelligence and Cognitive Science in the School of Computer Science at the University of Birmingham, UK.

The Centre for Applied Philosophy and Public Ethics (CAPPE) was established in 2000 as a Special Research Centre in applied philosophy funded by the Australian Research Council. It has combined the complementary strengths of two existing centres specialising in applied philosophy, namely the Centre for Philosophy and Public Issues (CPPI) at the University of Melbourne and the Centre for Professional and Applied Ethics at Charles Sturt University. It operates as a unified centre with two divisions: in Melbourne at the University (...) of Melbourne and in Canberra at Charles Sturt University. The Director of CAPPE and the head of the Canberra node is Professor Seumas Miller. Professor C.A.J. (Tony) Coady is the Deputy Director of CAPPE and the head of the Melbourne node. (shrink)

The broad range of capabilities exhibited by humans and animals is achieved through a large set of heterogeneous, tightly integrated cognitive mechanisms. To move artificial systems closer to such general-purpose intelligence we cannot avoid replicating some subset—quite possibly a substantial portion—of this large set. Progress in this direction requires that systems integration be taken more seriously as a fundamental research problem. In this paper I make the argument that intelligence must be studied holistically. I present key issues that must be (...) addressed in the area of integration and propose solutions for speeding up rate of progress towards more powerful, integrated A.I. systems, including (a) tools for building large, complex architectures, (b) a design methodology for building realtime A.I. systems and (c) methods for facilitating code sharing at the community level. (shrink)