G. Selfridge was born in London 10 May PANDEMONIUM: A PARADIGM FOR LEARNING O. G. SELFRIDGE INTRODUCTION WE are proposing here a model of a process which we claim can adaptively improve itself to handle certain pattern recognition problems Which cannot be adequately specified in advance. Such problems are usual when trying' to build a machine to Imitate any one of a very large class of human data processing techniques. A speech typewriter is a good example of something that very many people have been trying unsuccessfully to build for some time. We do not suggest that we have proposed a model which can learn to typewrite from merely hearing speech. Pandemonium does not, however, seem on paper to have the same kinds of inherent restrictions or inflexibility that many previous proposals have had. The basic motif behind our model is the Inn of parallel processing. This is suggested on two grounds: first, it is often easier to handle data in a parallel manner, and, indeed, it is usually the more "natural" manner to handle it in; and, secondly, it is easier to modify an assembly of quasi We are not going to apologize for a frequent use of anthropomorphic or biamorphic terminology.

Everything symbolized by a set of symbols constitutes the domain of symbolization of the set. The ultimate elements of the domain which symbolize nothing further are designated the terminal ind1catum. Most domains of symbolization comprise mediate symbols which are both symbolized by other symbols and themselves indicate further symbols. Mental concepts are treated as symbols. In translation, a set of symbols is transformed to another set in another language, the two sets having terminal indicata that only differ within narrow limits.

The Wageningen University & Research Library Catalogue contains a.o. the titles of the books and journals held by the participating libraries. Journal articles are not included in this database. Subjects covered include Agrotechnology, Food and Food Production, Plant and Animal Sciences, Soil Science, Geo-information, Landscape and Spatial Planning, Water and Climate, Ecosystem Studies, Economics and Society. The catalogue is updated daily and contains over 700.000 records.

Often they "work," but the results are variable and

Dr. Lucien Mehl, born 1919 in Paris, studied at the University, Paris where he obtained his degrees in Philosophy and Law, and a Diploma of Advanced Studies in Political Economy and at the National School of Administration. He is now'Maitre des Requetesi to the Council of State and Director of external training at the National School of Administration. He is a member of the International Fiscal Association, the International Cybernetics Association and the French Operational Research Society. He has published a number of articles on administrative science, law, cybernetics and operational research. LUCIEN HEEL INTRODUCTION I. It may seem an ambitious step to try to apply mechanization or automation to the legal sciences. However, a machine for processing information can be an effective aid in searching for sources of legal information, in developing legal argument, in preparing the decision of the administrator or judge, and finally in checking the coherence of solutions arrived at. In the first place, much preliminary work is needed for introducing automation in legal affairs, and so much work can only be decided upon if it is found to be of definite use. Secondly, such an undertaking is not without its risks; the jurist may lose direct contact with the sources of law and no longer have the benefit of the intellectual activity involved in searching for information. Lastly, as a result of mechanization of this kind, thought may itself become inflexible, diminishing creative power and innovative effort. Nowadays, however, machine processing of information is becoming essential; "Homo sapiens" is in fact exposed to the risk of being overwhelmed by the vast accumulation of knowledge. It is becoming increasingly difficult to gain access to the sources of ideas, and the researcher wastes valuable time and often intensive mental effort in detailed and unprofitable research, never being sure whether his investigations will be fruitful, or whether he will not bypass the essential information. Moreover, it happens that writers doing research in the same field of knowledge are unaware of one another's work; and besides this, the difficulty of finding the information required makes the researchers specialize still more. They find it hard to link up the different disciplines, because they are generally doomed to remain in ignorance of everything outside their on customary field of investigation. In legal matters, the number of laws and regulations and the scope of jurisprudence are growing on an alarming scale, and everyone is complaining about the situation - administrators and judges, as well as those dealt with under the law.

The suggestion is based on the similarity of behaviour of these formal systems and or animals. The design of classification computers is discussed in the first paper; the design of conditional probability computers Is discussed in a third paper (Uttley, 1958, ref. 15); in both papers working models are described. FUrther reference to these papers will be by date only. It is the aim of the present paper to consider whether the two principles might operate in nervous systems. Mere are four requirements for the principle of classification to operate in an area of a nervous system.

THE application of learning machines to process control is discussed. Three approaches to the design of learning machines are shown to have more in common than is immediately apparent. These are (1) based on the use of conditional probabilities, (2) suggested by the idea that biological learning is due to facilitation of synapses and (3) based on existing statistical theory dealing with the optimisation of operating conditions. Although the application of logical-type machines to process control involves formidable complexity, design principles are evolved here for a learning machine which deals with quantitative signal and depends for its operation on the computation of correlation coefficients.

This third symbol indicates the pathological symptoms about which we have no information at all.

I do so here with a diffidence that will need no explanation to anyone whose ideas have been on the move for ten years. I am only thankful - and a little surprised - to find as much as I do with which I - and apparently others - now agree.

I am quite ready to subscribe to the already mentioned slogan that "whatever a human being can do,an appropriate machine can do, too"; but I do this only because.I regard the slogan as utterly trivial. At the moment, I am not talking about what maohines could do in principle but only about what actually existing or blueprinted machines could do, and it Is with regard to these that I utter my definite opinions. If someone wishes to write sciencefiction about information-processing centres of the (undetermined) future, let him do so and I shall discuss it with him over a glass of beer and even offer some startling suggestions of my own. If he is interested in improving the literature search process today, I would strongly advise him to forget about mechanizing abstracting or indexing. May I add that it is with a good amount of sorrow that I have come to this conclusion which is quite counter, to my temperament and my convictions (never published) of a few years ago.