Results


The sharing of structure in theorem-proving programs

Classics

We introduce the concept of the value of an expression in a binding environment which we use to standardize clauses apart and share the structure of parents in representing the resolvent. Lists provide the most obvious and natural representation of literals because lists perfectly reflect function nesting structure. Therefore, we introduce the concepts of an expression, a binding environment, and the value of an expression in a binding environment. If VALI and VA L2 have no common instance, then the call will return false.


Social Implications of Intelligent Machines

Classics

Sociologists are concerned to predict the effect of changes on future society.But is prediction in principle possible when intelligence is involved? Ifintelligence is the production of novelty, accurate prediction might seem to bestrictly impossible. However this may be, it seems that the present troubleabout social prediction is simply that there are no adequate theoreticalmodels of societies. This means that politicians are almost powerless topredict, plan, or control, except with incredible errors. We find ourselves injust this position in trying to assess the implications of future intelligence.Machine Intelligence 6


Perception, picture processing and computers

Classics

The machines (usually digital computers) will classify simple shapes and printed letters and digits represented (by means of a suitable television scanner) as a matrix of I s and Os. Psychologists concerned with analysing complex behaviour often have to select an appropriate level of description. Let us confine our attention to three levels: words, phrases, sentences. Figure 1 shows a set of rules subdivided into numbered groups (1, 2, 3,..., 6).


LISP 1.5 Programmer's Manual

Classics

"The LISP language is designed primarily for symbolic data processing. It has been used for symbolic calculations in differential and integral calculus, electrical circuit theory, mathematical logic, game playing, and other fields of artificial intelligence.LISP is a formal mathematical language. It is therefore podsible to give a concise yet complete description of it. Such is the purpose of this first section of the manual. Other sections will describe ways of using LISP to advantage and will explain extensions of the language which make it a convenient programming system."The M.I.T. Press


A design for an understanding machine

Classics

It also maintains that all human meaning may be exhaustively represented in terms of readings on a practically infinite number of calibrated standards, or, alternatively, by elaborate constellations of readings on a very small number of "element" standards. The resolution of a polysemantic ambiguity, by whatever method of translation, ultimately consists of exploiting clues in the words, sentences or paragraphs of text that surround the polysemantic word, clues which make certain of its alternate meanings impossible, and, generally, leave only one of its meanings appropriate for that particular context. Any English speaking human, upon encountering a sentence containing both "bank" and one or more of these clue words, will use the clue word's semantic content, if necessary, to help resolve the meaning of "bank". From there the machine would be programmed to utilize clues in the words surrounding "bank" which might be helpful for deciding which of that word's two meanings was appropriate in this case.


Empirical Explorations of the Geometry-Theorem Proving Machine

Classics

The problem of theorem-proving is, in a sense, of a particularly simplenature. Once a sequence of expressions is found that passes the test for aproof of the theorem (such a test always exists), one may, so to speak,"close the book" on that problem, provided that no stipulations have beenmade concerning the elegance required of the proof. But, basing our estimateon the work of Newell, Shaw, and Simon (1957), any computerextant would require times of the order of a thousand years to prove a notuncommon ten-step geometry theorem by exhaustively developing sequencesuntil one emerged that passed the test for a proof. What is clearlycalled for is a technique for generating sequences with a much higher apriori probability of being the solution to the problem than those generatedby an exhaustion algorithm. Proceedings of the Western Joint Computer Conference, 1960, 17:143-147.


Truth and probability

Classics

Footnote numbering is maintained as in the original text -- as a result, page numbers are also noted for where the footnote originally appears. Some minor typographical errors remain in the1931 edition which are corrected here: - - Page 169, line 19, we replaced "intensities fo feelings" with "intensities of feelings" Page 176, line 22 we replaced "few asumptions as possible" with "few assumptions as possible" - - Page 187, line 12, we replaced "objective interpetation" with "objective interpretation" Page 191, Footnote 1, we replaced "guided only be ratiocination" with "guided only by ratiocination". Keynes's symbolism p/h meaning the probability of proposition p given proposition h. "Truth and Probability" written 1926. CONTENTS (1) The Frequency Theory (2) Mr Keynes' Theory (3) Degrees of Belief (4) The Logic of Consistency (5) The Logic of Truth (1) THE FREQUENCY THEORY In the hope of avoiding some purely verbal controversies, I propose to begin by making some admissions in favour of the frequency theory.