@inproceedings{10.5555_3017065.3017066_, title = {How One Learns Strategies}, subtitle = {Processes and Representation of Strategy Acquisition}, author = {Anzai, Yuichiro}, year = {1978}, tags = {}, abstract = {A general theory is presented for the process and representation of strategy acquisition by human and artificial intelligence systems. The process is characterized by inductive generalization and organization of new rules through successive construction of new problem spaces. Various subtasks involved in the process can be naturally represented by subroutines in a hierarchical adaptive production system (HAPS), which is proposed as a representational scheme for strategy acquisition. HAPS has capabilities for subroutine calls and creating new productions in any subroutine. Results from strategy acquisition in the Tower of Hanoi puzzle by a HAPS program are given.}, keywords = {production systems, problem solving, strategy acquisition, rule learning}, numpages = {14}, pages = {1-14}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017067_, title = {Experience with a Refinement Paradigm in a Knowledge-Based Automatic Programming System}, author = {Barstow, David R.}, year = {1978}, tags = {}, abstract = {PECOS is a knowledge-based system that constructs concrete implementations of abstract algorithms in the domain of symbolic programming. The rules in PECOS's knowledge base deal explicitly with many details of symbolic programming, including intermediate-level constructs and certain design decisions. Programs are synthesized by gradually refining the original specification into a program in the target language. By applying the refinement rules in its knowledge base, PECOS constructs a refinement tree: the root is the abstract specification, each leaf is an alternative implementation, and each path is a different sequence of rule applications. Thus, PECOS can construct a variety of different implementations for the same abstract specification.}, numpages = {7}, pages = {15-21}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017068_, title = {On Strategies for the Synthesis of Algorithms}, author = {Bibel, Wolfgang}, year = {1978}, tags = {}, abstract = {A number of strategies for the construction of algorithms from a given input-output specification of a problem are presented, Their application is illustrated by a detailed synthesis of the usual MAXIMUM algorithm and Hoares FIND algorithm.}, numpages = {6}, pages = {22-27}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017069_, title = {A (Self-Adapting) Network for Recognition of Visual Structure}, author = {Birch, P.}, year = {1978}, tags = {}, abstract = {A system is presented that recognises emerging scene structures as a picture is interpreted. Emphasis is placed on using local gross features (which are cheap to compute) to guide the building of an interpretation. When this results in ambiguity, more global considerations can be used. Finer features (more expensive to compute) can also be utilised when present.}, keywords = {self-adapting, alternative interpretations, ambiguities, intermediate descriptions, 'Gestalt', perception, recognition}, numpages = {5}, pages = {28-32}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017070_, title = {Dialogues in Actor Nets}, author = {Böhm, H. P. and Fischer, H. L. and Raulefs, P.}, year = {1978}, tags = {}, abstract = {The actor society representation of knowledge is introduced. In a self-organizing society of actors, knowledge is distributed s.t. each individual actor has strictly local information. Although each actor exclusively tries to achieve its own private goal, it exhibits a communication behavior making the entire society form a well-orgaized system having global properties no individual actor is aware of. Actor societies are realized as networks of CSSA-actors /1/.}, numpages = {2}, pages = {33-34}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017071_, title = {The Fit Approach to AI Languages}, author = {Boley, Harold}, year = {1978}, tags = {}, abstract = {An overview of the AI language FIT is given, which attempts to unify, generalize, and advance a number of programming language concepts. Its central entities are fitters, a unification of functions, patterns, and variables, especially of predicate functions and typed variables. The latter are also viewed as new kinds of demons. Fitter definitions are assignments of bodies to complex variables viewable both as function application forms and as consequent theorem invocation patterns. Fitter computation and look-up are thus unified to fitter valuation.}, keywords = {FIT, pattern matching, demons, types, complex variables, valuation, fitments, fitters, consequent theorems, AI languages}, numpages = {7}, pages = {35-41}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017072_, title = {Dangerous Behaviour}, author = {Brady, J. M.}, year = {1978}, tags = {}, numpages = {4}, pages = {42-45}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017073_, title = {Experimental Learning Model}, author = {Brazdil, Pavel}, year = {1978}, tags = {}, abstract = {The aim of our work is to investigate how a relatively small set of rules can be transformed into a running program capable of solving a fairly wide variety of problems. We have written a program ELM (Experimental Learning Model) which modifies a given set of rules (clauses) as a result of its experience. The problems are chosen from the domain of school arithmetic/algebra and solved by one part of our system (with the help of the teacher). The solutions are analyzed and modifications to the existing system of rules are performed as a result.}, numpages = {5}, pages = {46-50}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017074_, title = {Understanding the Image Forming Process}, author = {Brooks, M. J.}, year = {1978}, tags = {}, abstract = {Relatively little work has been done to try to understand the image forming process. It is contended here that this situation must change if vision programs are to attain any level of sophistication. Some difficult problems are posed and, as a first step towards answering them, an attempt is made to understand the effects of a light-source which is a 'sky' rather than the traditionally adopted point source.}, keywords = {point and sky light sources, analytic approach, image forming process, reflectivity function}, numpages = {5}, pages = {51-55}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017075_, title = {A Sequent Cadculus for Modal Quantificational Logic}, author = {Brown, F. M.}, year = {1978}, tags = {}, abstract = {We give an axiomization of a modal quantificational logic which captures the notion of logical truth. This modal logic is stronger than S5. Next we describe a sequent calculus for this modal logic, and show that it is complete. Finally we describe some potential applications of Modal Logic to Artificial Intelligence.}, numpages = {10}, pages = {56-65}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017076_, title = {Analysing and Representing Natural Language in Logic}, author = {Brown, F. M. and Schwind, C. B.}, year = {1978}, tags = {}, abstract = {We-describe the basic idea of a theory of natural language understanding in which the meaning of natural language text and the laws for natural language analysis are both represented in logic.}, numpages = {5}, pages = {66-70}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017077_, title = {Knowledge about Knowledge}, subtitle = {Making Decisions in Mechanics Problem Solving}, author = {Bundy, Alan and Luger, George and Mellish, Chris and Palmer, Martha}, year = {1978}, tags = {}, abstract = {In October 1975 we started a three year SRC supported project on "A Program to Solve Mechanics Problems Stated in English". This project is now beginning its third year and many of its initial objectives have been achieved. In this paper we briefly review the progress that has been made and discuss some of the more promising areas of research that we have uncovered.}, numpages = {11}, pages = {71-81}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017078_, title = {A Theorem Prover Dating a Semantic Network}, author = {Champeaux, Dennis de}, year = {1978}, tags = {}, abstract = {Ordinary predicate calculus (p.c.) is equiped with sorts, an extension of types, allowing a concise formulation of problems. A semantic network can easily provide the sort-part of the input of a theorem prover based on a sorted p.c. Experimental results are reported of such a theorem prover.}, keywords = {typed predicate calculus, n1-resolution, sorts, semantic networks, theorem proving}, numpages = {11}, pages = {82-92}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017079_, title = {Determining the Orientation of Surfaces from Optical Flow}, author = {Clocksin, William F.}, year = {1978}, tags = {}, abstract = {This paper presents a method for computing the orientation of physical surfaces relative to the moving observer. The method is theoretically derived; it is not 'ad hoc'. The method exploits the remarkable properties of the optical flow, a concept which figures prominently in the Gibsonian theory of visual perception. We first describe the dynamic nature of visual information, and formally derive the optically sensed velocity field transformation (optical flow) for a moving observer. Then we derive the inverse transformation to show how gradients of optical flow uniquely inform about the orientation and shape of physical surfaces.}, keywords = {physical constraints, optical flow, motion perception, motion contrast, surface orientation, optically sensed velocity fields}, numpages = {10}, pages = {93-102}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017080_, title = {Problem Solving by Hyperplanning}, author = {Eick, Ch. and Raulefs, P.}, year = {1978}, tags = {}, abstract = {Almost all current problem-solvers are unable to improve their behavior by analysing failures and alter their planning strategies. Only HACKER /1/ provides just a few error-analysis and error-correction mechanisms, but there is no systematic way of incorporating domain-specific knowledge.A hyperplan consists of a sequence of operators and meta-operators both linked to various kinds of nodes for book-keeping. Operators perform state-transformations specified by conditional add/delete sets. Meta-operators are capable of transforming the hyperplan they are part of. Specific conflict situations spontaneously activate conflict administrators that use domain-specific criteria to select appropriate conflict operators for resolving conflicts. Both conflict and meta-operators can analyse a hyperplan and make appropriate corrections to avoid subsequent conflicts in similar situations.}, numpages = {2}, pages = {103-104}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017081_, title = {Paramodulated Connection Graphs}, author = {Eisinger, Norbert and Siekmann, Jörg and Wrightson, Graham}, year = {1978}, tags = {}, numpages = {5}, pages = {105-109}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017082_, title = {A System for Visual-like Understanding of LISP Programs}, author = {Goossens, Daniel}, year = {1978}, tags = {}, abstract = {Presented is an implemented systems CAN, which understands LISP programs. The system is a program debugging tool and is intended to use procedural knowledge about concrete list structures. The set of rules which implement this knowledge is represented as a unification process, called META-PATTERN-MATCHING, which handles "constrained segment variables".Dealing with the "visual aspects" of data structures, the system CAN is useful as the programmers' assistant.}, keywords = {unification, meta-interpretation, meta-pattern-matching, generalisation, propagation, understanding}, numpages = {5}, pages = {110-114}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017083_, title = {A Method for Computing Heuristics in Problem-Solving}, author = {Guida, Giovanni and Somalvico, Marco}, year = {1978}, tags = {}, abstract = {The purpose of this paper is to propose a formal framework for structuring and embedding the heuristic information, in order to allow an algorithmic computation, in quite general cases, of the evaluation function undefined (n) of the classical Hart-Nilsson-Raphael algorithm, The notion of semantic graph is first introduced, in which the atomic notion of node is expanded by associating to it an internal structure where the heuristic information is inserted. It is proved that undefined(n) can be computed by solving an auxiliary problem, obtained from the original one, by adding new arcs, and of smaller complexity than that one. A class of algorithms for the computation of undefined(n) is then defined. The validity of the model proposed is discussed in detail.}, numpages = {7}, pages = {115-121}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017084_, title = {HAM-RPM}, subtitle = {Natural Dialogues with an Artificial Partner}, author = {Hahn, W. v. and Hoeppner, W. and Jameson, A. and Wahlster, W.}, year = {1978}, tags = {}, abstract = {This paper introduces the understanding system HAM-RPM, which simulates a dialogue partner conversing about a real-life domain. After outlining the system's overall structure, we discuss three of its distinguishing features: The first is its organization of spatial data in a redundant multiple data base, inspired by certain aspects characteristic of visual search in humans. A new algorithm for noun-phrase generation is then sketched which is sensitive to the conversational state and uses a 'worst-case-first' strategy. Finally, we describe in some detail a specifie operationalisation of the notion of the communicative relevance of objects. The paper concludes with a summary of the objectives of this research.}, keywords = {natural language processing, focus of attention, dialogue simulation, fuzzy, visual search, reference semantics}, numpages = {10}, pages = {122-131}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017085_, title = {A Mutiprocessor Systm for Kowledge Representation and Manipulation}, author = {Hanna, F. K. and Harwood, W. T.}, year = {1978}, tags = {}, abstract = {Knowledge manipulation tasks are computationally intensive : one way to provide sufficient computing power is to use a loosely coupled network of microprocessors. For this approach to be efficient it is necessary that the overall task be divisible into parts which can be run in parallel and which are, to a degree, self-contained. It is argued that contemporary approaches to knowledge manipulation systems fulfil these criteria. The paper describes the design of such a system, currently under construction. It is programmed in a LISP-like language which incorporates parallelism, and allows programs to be run in remote environments. Aspects of this language, its implementation and the widerlying hardware are described.}, keywords = {parallelism, knowledge representation, microprocessors, LISP, multiprocessor system}, numpages = {6}, pages = {132-137}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017086_, title = {Program Synthesis in Predicate Logic}, author = {Hogger, C. J.}, year = {1978}, tags = {}, abstract = {Research in automatic theorem proving has led to an interpretation of predicate logic as a high-level procedural programming language. Logic 'programs' can be executed by interpreters based on the resolution principle. Logic offers a single formalism in which programs can be specified, derived and finally expressed using the same tool (theorem proving) as is used for executing them. This paper describes the derivation of logic programs from specifications written in standard logic.}, numpages = {9}, pages = {138-146}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017087_, title = {Vision}, author = {Horn, Berthold K. P.}, year = {1978}, tags = {}, numpages = {17}, pages = {147-163}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017088_, title = {Ani}, subtitle = {Am Example of Computational Creativity}, author = {Kahn, Kenneth M.}, year = {1978}, tags = {}, abstract = {A computer program, named Ani is presented in this paper.1 When completed Ani will be capable of producing non-representational narrative computer animation in response to very high-level incomplete descriptions. Ani brings together many different sources of knowledge in the form of suggestions which interact and are combined in complex ways. Initially Ani is faced with a very under-constrained situation. As choices are made constraints are added in soch a way to keep the film both coherent and possible. This system is presented as one example of a computational process of creativity.}, numpages = {5}, pages = {164-168}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017089_, title = {A Sane Algorithm for the Synthesis of LISP Functions from Example Problems}, subtitle = {The "Boyer and Moore Algorithm"}, author = {Kodratoff, Yves and Fargues, Jean}, year = {1978}, tags = {}, abstract = {We study the problem of synthesis of the algorithmic representation of a LISP function f, given its domain and a finite subset of its graph. The solution is an interpretation of a program scheme. Instanciation of the scheme variables is made by a new method which uses subsumption and then, if the subsumption fails, uses generalization so that, at a further step, the subsumption of the generalized expressions succeed.}, keywords = {difference equations, program synthesis, generalisation, LISP programs, subsumption, pattern matching, program schemes}, numpages = {7}, pages = {169-175}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017090_, title = {A Study in the Application of Theorem Proving}, author = {Luckham, David C. and Morales, Jorge J. and Schreiber, Joachim F.}, year = {1978}, tags = {}, abstract = {With selected examples we illustrate some useful applications of theorem proving, which have been considered during the last few years, and show, at the same time, the difficulties which have arisen. We aim at some kind of "hunch" language to guide in the generation of clauses. This paper is a short account of an unpublished manuscript by the same authors.}, keywords = {answer extraction, application of theorem proving, interactive theorem proving, choice strategy, program verifications predicate logic programming, exploratory use, edit strategy, proof checking}, numpages = {13}, pages = {176-188}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017091_, title = {A Quasi-Parallel Contour Following Algorithm}, author = {Mérö, L.}, year = {1978}, tags = {}, abstract = {A procedure for extracting the contour lines and arcs from digitized TV pictures is described. Initially the picture is subjected to an edge detection operation resulting in short edge-line pieces /strokes/. The bifurcation points /corners/ of the contours are detected based on the variances of the angles of the strokes.Mathematical criteria are stated on the arrays of strokes /streaks/ to be found among the corners in order to approximate the contours optimally. An algorithm is given which finds the streaks satisfying these criteria.Any number of processors can work simultaneously in performing the algorithms, therefore the algorithm is called quasi-parallel.}, keywords = {picture processing, corner finding, parallel algorithms, contour following, minimal cost path}, numpages = {5}, pages = {189-193}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017092_, title = {Advice Table Representations of Chess End-Game Knowledge}, author = {Michie, Donald and Bratko, I.}, year = {1978}, tags = {}, abstract = {Computer implementation is described of the chess ending king and knight versus king and rook, and-the trade-off between search and domain-specific knowledge is examined.}, numpages = {7}, pages = {194-200}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017093_, title = {Representation of Phonetic and Phonemic Knowledge in a Speech Understanding System}, author = {Mori, Renato De and Laface, Pietro}, year = {1978}, tags = {}, abstract = {The paper describes the organization of knowledge source and the algorithms for evaluating hypotheses of the front-end acoustic classifier of a Speech Understanding System.}, numpages = {5}, pages = {201-205}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017094_, title = {The Last Survey of Representation of Knowledge}, author = {McDermott, Drew}, year = {1978}, tags = {}, abstract = {This is a survey of recent progress in "representation of knowledge." First the field is defined. Then progress in general techniques is surveyed. Then a catalogue of results in particular domains is given. Finally, a few recommend ations are listed.}, numpages = {16}, pages = {206-221}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017095_, title = {Best-First Interpretation of Imperfect Object Contours}, author = {Neumann, Bernd}, year = {1978}, tags = {}, abstract = {A scene-analysis system is presented which identifies objects by means of straight-line contour approximations. Descriptions of prototypes are stored in a relational-model database. Scenes may be disturbed and may contain objects which are partially occluded by another object. A segmentation algorithm extracts straight edges by multidirectional sampling of the picture gradient matrix. Interpretation hypotheses are generated using 2-step relational matching. A best-first strategy selects the final interpretation using constraint filtering for ambiguous situations. Results are presented for scenes containing configurations of artificial shapes (card-board pieces) and real-life objects (tools). They indicate that in systems designed for application one should sophisticate interpretation rather than segmentation.}, keywords = {constraint filtering, occlusion, scene analysis, interpretation, contour approximation}, numpages = {7}, pages = {222-228}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017096_, title = {A Production Rule Account of Errors in Children's Subtraction}, author = {O'Shea, Tim and Young, Richard M.}, year = {1978}, tags = {}, abstract = {Many of the errors that occur in children's subtraction are due to the use of incorrect strategies rather than to the incorrect recall of number facts, A production system is presented for carrying out two- column subtraction using the method of decomposition. We show that many of the incorrect strategies used by schoolchildren can be modelled as the consequences of simple changes to this production system, such as the omission of individual rules or the addition of rules appropriate for other arithmetical tasks.}, numpages = {9}, pages = {229-237}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017097_, title = {A Philosophy of 'wicked' Problem Implementation}, author = {Partridge, Derek}, year = {1978}, tags = {}, abstract = {An important characteristic of intelligence is the ability to deal adequately with 'wicked' or ill-structured problems. A methodology is proposed for the computer implementation of certain classes of 'wicked' problems. Curiously, this methodology contrasts sharply with the current artificial intelligence paradigm: a rationalization of this disparity is given. Lastly, the programming methodologies of software engineering appear to be most closely akin to the current inquiry, and this contention is examined.}, numpages = {10}, pages = {238-247}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017099_, title = {A Network Representation for Procedural Knowledge}, author = {Phillips, Brian}, year = {1978}, tags = {}, abstract = {Two kinds of knowledge necessary for text understanding systems that are procedural, not deciarative, are processes on the knowledge structure (e.g., path-tracing) and knowledge of concepts like "analogy" and "euphemism". The desirability of being able to manipulate procedures as data makes a case for having the same format for both declarative and procedural knowledge. This is achieved by writing procedures in a semantic net formalism. Executing such a procedure produces a network that explicitly contains the control structure as well as the data.}, keywords = {semantic networks, procedural knowledge}, numpages = {5}, pages = {253-257}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017100_, title = {A Computational Study of a Period of Infant Development}, author = {Prazdny, K.}, year = {1978}, tags = {}, abstract = {A period in the early human development ranging from about 12 to 20 weeks is described in terms of typical behaviours characteristic of this developmental stage, The behaviours are reassessed in terms of a computer model and it is argued that they are a result of the interaction between perceptual and conceptual levels of the system and the way in which a conflict between competing descriptions of the object attended to is resolved. As such,the rationale underlying the program suggests that it is the object and its event description which form the basic unit of analysis and not the infant's previous activity as argued by Piaget (Piaget, 1954), or the motion or places as argued by Bower (Bower, 1971).}, keywords = {object identity, motion perception, object concept development, developmental model}, numpages = {5}, pages = {258-262}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017101_, title = {Parameterized Region Extraction for the Description of Moving Objects}, author = {Radig, Bernd}, year = {1978}, tags = {}, abstract = {A sequence of TV-frames is recorded from a real-world scene. In order to determine motion attributes of objects the segmentation of these (digitized) TV-frames has to provide a reliable symbolic description for each image. The complexity of light intensity distribution along object surfaces in such scenes calls for a method which can extract regions according to various criteria. The algorithm presented here transforms the grayvalue matrix into a vector field, the vector components describing small local planes fitted to the grayvalue image. A three-level filter process uses parameterized constraints to remove inconsistent nodes and arcs from a graph which represents the vector field. Nodes are associated with vectors, arcs are initialized to connect adjacent vectors. A region is associated with an isolated graph component, its features are computed from the vectors in the component. Comparison of features establishes correspondence of regions originating from the same object surface in different frames. These region sequences serve as a basis for object description, especially its motion state.}, numpages = {10}, pages = {263-272}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017102_, title = {Predictions and Procedures in Semantically-Based Grammar}, author = {Ritchie, Graeme D}, year = {1978}, tags = {}, abstract = {It has been suggested that grammars for analysing English sentences should be based on semantic constructs. We examine what this might mean, focussing on the questions of semantic predictions and of processing procedures. There seems to be no general way of organising processing units around semantic categories, or of relating semantic predictions to specific processing units. We conclude that processing rules cannot be expressed in terms of referential semantics, and that top-down predictions can do little to drive the actual processing of words.}, numpages = {10}, pages = {273-282}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017103_, title = {What's in a Social Action for U?}, author = {Rosner, M.}, year = {1978}, tags = {}, abstract = {What is a social action, and what would it be for a social actions to be incorporated into a natural language understanding system? A simple notation for talking about the goals of the participants is introduced and it is suggested that given a sufficiently precise semantics, this notation could efficiently describe the knowledge needed by a system which could use, and understand the use by others, of a small subset of social actions.}, keywords = {action, politeness, dialogues, social action, notation, conversation}, numpages = {10}, pages = {283-292}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017104_, title = {A Conversational System Which Understands Short Stories}, author = {Sabah, G.}, year = {1978}, tags = {}, abstract = {We present a conversational system which can learn short stories in a natural language and perform some reasoning about them.In our model, a sentence is represented by a purely semantic case structure ("deep structure") with embedding properties.The vocabulary is a tree of word representative elements, each of which having its sense precised by some semantic features (pointers to other words). To understand a sentence within a given story, we reduce the information it carries, This reduction gives way to the application of a few rules for character behaviour, whose aim is to connect the sentence to another onp by establishing some causal relationship between desires and actions. The system can then answer "why" questions and questions about some facts not explicitly described in the story.The natural language used is French, the program is written in SIMULA}, numpages = {5}, pages = {293-297}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017105_, title = {What Structure Does a LISP Program Have - besides Being a Set of Procedures?}, author = {Sandewall, Erik}, year = {1978}, tags = {}, abstract = {The hierarchical structure of algol-like languages is well known. In procedure-oriented languages such as APL or LISP, a program consists of a large number of procedures, each of which is usually small in size. For large programs, consisting of hundreds of procedures, it is necessary to have some higher-level structure to organize the procedures. The present paper describes such a higher-level structure. A programmer's-assistant-type program which supports the model has been implemented, and the experience in using this program gives some support for the model.}, numpages = {6}, pages = {298-303}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017106_, title = {Representing Actions by State Logic}, author = {Schwind, Camilla B.}, year = {1978}, tags = {}, abstract = {The following paper presents a formal calculus interpreted by Kripke-type semantics for the description of the actions of an intelligent system, The frame problem is reduced to the logical problem of proving consistency.}, numpages = {5}, pages = {304-308}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017107_, title = {Representation and Control in Vision}, author = {Sloman, Aaron and Owen, David and Hinton, Geoffrey and Birch, Frank and O'Gorman, Frank}, year = {1978}, tags = {}, numpages = {6}, pages = {309-314}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017108_, title = {Design for a Plan Hypothesizer}, author = {Sridharan, M. S. and Smith, D.}, year = {1978}, tags = {}, numpages = {10}, pages = {315-324}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017109_, title = {The Frame Shift Mechanism in Natural Language}, author = {Steels, Luc}, year = {1978}, tags = {}, abstract = {The frame shift mechanism is a method to resolve the conflict between the goals of a certain language communication and the tools available for turning these types of goals into natural language. It consists in assembling a new specification relating cognitive goals with natural language means.}, numpages = {5}, pages = {325-329}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017110_, title = {Planning in Uncertain Environments through Situation Calculus}, author = {Štĕpánková, Olga}, year = {1978}, tags = {}, abstract = {We try to broaden the scope of the situation ealeulus developed in the tradition of Green /69/, Mc Carthy, Hayes /69/ to cope with uncertainity. The situation dependent object functions are used for that purpose. Our method is explained on examples of "cat and mouse in a maze" and a BRIDGE termination. We shall impose some restrictions on the world, the robot and its goals under which a theorem prover can carry the planning of robot a activity in a changing or not completely known environment. Finally we show how the activity of the executing robot of Siklossy can be planned in our formalism.}, numpages = {10}, pages = {330-339}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017111_, title = {Completeness and Soundness of the Connection Graph Proof Procedure}, author = {Stephan, W. and Siekmann, J.}, year = {1978}, tags = {}, numpages = {5}, pages = {340-344}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017112_, title = {Common-Sense Representation}, author = {Suzman, P. L.}, year = {1978}, tags = {}, abstract = {Some problems with existing formal representation languages are discussed, and an alternative semantics is proposed that attempts to build in a particular view of the world. This is essentially that the world consists of objects whose properties are determined by the properties of their parts. The semantics also makes provision for predicates to have intensions as well as extensions, so they can be reasoned about. The implications of the semantics are discussed with particular reference to the problems of representing change, where it is contrasted to the approach taken in STRIPS and the Situational Calculus.}, keywords = {semantics, representational language, representation of change}, numpages = {6}, pages = {345-350}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017113_, title = {Toward Greater Generality in Artificial Intelligence}, author = {Uhr, Leonard and Kochen, Manfred}, year = {1978}, tags = {}, abstract = {We have been asked to discuss how research in Artificial Intelligence and Cognitive Modelling might best progress, and to comment on this meeting's papers. The comments must wait for the oral presentation. Here we examine two major issues: A) We describe an approach toward more general AI systems, and pose some rather simple-seeming yet unsolved touchstone problems to be attacked. B) We suggest that AI should attack our very difficult problems with as diverse as possible a set of good approaches.}, numpages = {4}, pages = {351-354}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017114_, title = {AI Programming Methodology}, author = {Wielinga, Bob J.}, year = {1978}, tags = {}, numpages = {20}, pages = {355-374}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @inproceedings{10.5555_3017065.3017115_, title = {Grammar Discovery as Data Compression}, author = {Wolff, J. Gerard}, year = {1978}, tags = {}, abstract = {A discovery procedure is described for non-recursive context-free phrase structure grammars which is based on three data compression principles. It works without the need for semantic information, negative samples, a teacher or pre-segmented data.}, numpages = {5}, pages = {375-379}, crossref = {78:_proceed_aisb_gi_confer_artif_intel_}, } @proceedings{78:_proceed_aisb_gi_confer_artif_intel_, title = {Proceedings of the AISB/GI Conference on Artifcial Intelligence}, editor = {Bundy, Alan and Hardy, Steve and Nagel, H.H. and Pitrat, Jacques and Sleeman, Derek and Wilks, Yorick}, year = {1978}, tags = {}, country = {Hamburg}, pdf_locked = {False}, url = {https://dl.acm.org/doi/proceedings/10.5555/3017065}, url2 = {https://aisb.org.uk/wp-content/uploads/2019/12/aisb1978.pdf}, file = {/media/john/data/todo/pdfs/proceedings/aisb/aisb1978.pdf}, }