Terry Winograd, A Procedural Model of Language Understanding, Computer Models of Thought and Language, R. Schank and K. Colby, eds., 152-186, New York: W.H. Freeman, 1973.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....since it relies on the full power of interpretation of the sensed data describing the environment in order to allow the symbolic reasoning processes to be applied to the diagnosis and correction of these failures. A similar approach, although in a different context, was used in the Shrdlu system [Win73], in which the Shrdlu system planned and executed simple blocks world tasks. In this case, however, the world was simulated and therefore the linkage between actions and execution instructions was not subject to failure as it is in interaction with a non simulated environment. Of course, it is ....

T.Winograd, A procedural model of language understanding, R.C.Schank and K.M.Colby (eds), Computer Models of Thought and Language, Freeman, 1973. 10

.... only for cognitive science [Brown and Levinson, 1987] but even for human computer dialogue [Cassell and Bickmore, 2002] 3 Computational researchers interest in task oriented dialogue has persisted from early work on domain specific question answering systems [Green et al. 1961, Woods, 1968, Winograd, 1973] to such current efforts in spoken dialogue agents as [Allen et al. 1995, Ferguson and Allen, 1998, Wahlster, 2000] 2 e A: OK, I ll take care of it. 1) suggests the effort that people make when they collaborate to maintain a detailed shared understanding of the status and direction of their ....

Winograd, T. (1973). A procedural model of language understanding. In Schank, R. and Colby, K., editors, Computer Models of Thought and Language, pages 152--186. W. H. Freeman. Reprinted in Grosz et al. (eds), Readings in Natural Language Processing.

....demand that concepts are represented by a single biological neuron, but allows for a small, functionally distinct cluster of neurons. CHAPTER 2. SETTING THE STAGE 18 This architecture could be considered an implementation of learning procedural semantics, as pioneered by Winograd s SHRDLU system #Winograd 1973#. SHRDLU was a question answering system which operated in the time honored blocks world domain. The user could, for example, request that the system #Pick up the red block, and the appropriate object identi#cation and action would occur. If the request was ambiguous, the system might reply ....

Winograd, Terry. 1973. A procedural model of language understanding. In Computer Models of Thought and Language.NewYork: W. H. Freeman.

....discourse game can be designed which does justice to the famed triad of thesis, antithesis and synthesis. Appendix A The Article Nine World Just as Winograd s SHRDLU system conversed in the famous blocks world , variants of which have continued to be of great service in AI planning research [129], the examples used in this thesis are based, not on the real Article Nine, with all its complexity, but on a micro world version of it. Let s call it the Article Nine World , or simply A9W . Our goal is a simplified version of Article Nine which, although certainly inadequate for solving real ....

Terry Winograd. A Procedural Model of Language Understanding. In R. C. Schank and K. M. Colby, editors, Computer Models of Thought and Language. Freeman, San Francisco, 1973.

....of Figure 1. For example, if it learns that shove labels pushing actions with high acceleration, it must be able to carry out a shoving action when asked. This is done byinterpreting a verbal command as choosing a motor action and its parameters in accordance with 1 The project is reminiscent of Winograd s #1973# shrdlu, although it di#ers in its focus on learning, its restriction to verbs, and its attention to #ner semantic distinctions. Siskind #1992# has considered action verb learning but does not leverage internal state #e.g. intentions, motor commands#, relying instead on visual features alone. ....

Winograd, T. #1973#. A procedural model of language understanding. In Computer Models of Thought and Language.W.H.Freeman, New York.

....and exhibit them through the use of computers. This inundation is further proof that we have not quite reached our goal and the need to learn more is ongoing. An early attempt at defining a model for language understanding and developing a system to test it is illustrated in [Winograd, 1972] and [Winograd, 1973]. His focus is on the ability to write a computer program that can perform some semblance of natural language understanding and that in the process it may help clarify what language is, how we use it to communicate, and more importantly how it may bring us a step closer to understanding human ....

Terry Winograd. A procedural model of language understanding. In Computer Models of Thought and Language. W.H. Freeman and Company, 1973.

....to existing research. Throughout, the key issue addressed is the correspondence problem, namely how to associate visual events with words and vice versa. 1 Introduction Much has been said about the necessity of linking language and vision in order for a system to exhibit intelligent behaviour [Win73, Wal81] A complete natural language understanding system should be able to understand references to the visual world, especially if it is engaged in discourse or conversation or even reading narratives. Without the ability to visualise, a discourse understanding system does not have access to a ....

....as diagrams) text and pictures need to be considered as a whole when it comes to understanding. It described the need for a computer system which could deal with both types of input using uniform methods. The paper is now best known for introducing the idea of syntactic pattern recognition. Win73] describes one of the first systems to attempt an integration of language and vision, by accepting block manipulation instructions in English and displaying the results visually. The ultimate goal was to develop a system for natural language conversation which could incorporate visual information ....

T. Winograd. A Procedural Model of Language Understanding. In Computer Models of Thought and Language, pages 152--186. W. H. Freeman and Company, San Fransisco, 1973. 23

....learn to shift between levels, learning which levels are best for which situations. Resnick and Martin 1990) 12. Ideally, this would include some notion of episodic memory. Unfortunately, only two major examples of agents incorporating episodic memory in the literature easily come to mind: Winograd s (1973) SHRDLU and Vere and Bickmore s (1990) basic agent. For a thought provoking look into the consequences of a future where a personal agent might become the ultimate cradle to grave companion, experiencing and remembering every event of a lifetime, see The Teddy chapter in Norman (1992) 13. ....

Winograd, T. 1973. A Procedural Model of Language Understanding. In Computer Models of Thought and Language, eds. R. Schank and K. Colby, 249--266. New York: Freeman.

....(numberp x) x 62 N Chapter 2 Plan Verification in the Blocks World In this chapter, we use our method of formalizing problem domains in the Boyer Moore logic to specify the blocks world described in Chapter 1. The blocks world has been used as an example problem domain in AI for a long time [118, 112, 28] and at present there is some controversy over whether or not it is a solved problem. It is therefore worth emphasizing that we are interested in an implementation that would allow us to express all possible problems and plans, including those that involve conditionals and recursion such as our ....

....two different representations of blocks world states. Our first representation of blocks world states is as a list of terms that stand for primitive relations between blocks in a state such as On(A; B) Ontable(D) and Clear(C) This representation has been used in AI programs written in PLANNER [118] and Prolog [58] and is used currently by many existing planning programs [113] The representation turns out to be inconvenient for specifying the predicate on the set of possible states because it forces us to specify a number of state constraints explicitly. By choosing a different ....

T. Winograd. A procedural model of language understanding. In R. Schank and K. Colby, editors, Computer Models of Thought and Language. W. H. Freeman, San Francisco, 1973.

....1957] 6) Colorless green ideas sleep furiously. On the other hand, people can put up with imperfect syntax and get the meaning out of ungrammatical strings of words. For instance, even in the total absence of syntax as in (7) we still get some meaning out of the text. 7a) Skid crash hospital. [Winograd, 1973] (7b) Fire match arson hotel. Charniak, 1983] Lack of independence of syntax also leads to difficulties in accounting for syntactic generalizations across different semantic entities. An example of a syntactic generalization exhibited by the language processor in a variety of constructs is the ....

T. Winograd. A Procedural Model of Language Understanding. In R. C. Schank and K. M. Colby, editors, Computer models of thought and language, pages 152--186. W. H. Freeman, 1973.

....demand that concepts are represented by a single biological neuron, but allows for a small, functionally distinct cluster of neurons. CHAPTER 2. SETTING THE STAGE 18 This architecture could be considered an implementation of learning procedural semantics, as pioneered by Winograd s SHRDLU system (Winograd 1973). SHRDLU was a question answering system which operated in the time honored blocks world domain. The user could, for example, request that the system Pick up the red block, and the appropriate object identification and action would occur. If the request was ambiguous, the system might reply ....

Winograd, Terry. 1973. A procedural model of language understanding. In Computer Models of Thought and Language. New York: W. H. Freeman.

....green ideas sleep furiously. On the other hand, people can put up with imperfect syntax and get the meaning out of ungrammatical strings of words. For instance, even in the (nearly) total absence of syntax as in (11) we still get some meaning out of the text. 30 (11a) skid crash hospital (Winograd, 1973) (11b) fire match arson hotel (Charniak, 1983) Further support for syntactic autonomy comes from the complexity of interaction and that of integrating preferences from several higher levels in making on line decisions. The speed and automaticity of decisions at the lower levels of sentence ....

....AND SENTENCE PROCESSING ARCHITECTURES What really seems to be going on is a coordinated process in which a variety of syntactic and semantic information can be relevant, and in which the hearer takes advantage of whatever is more useful in understanding a given part of a sentence. T. Winograd, 1973 In the previous chapter, we identified the subtasks of syntactic and semantic analyses and showed how they are necessary for both word meaning selection and word meaning composition. We also noted that, in general, neither syntactic nor semantic analysis can be completed without useful ....

[Article contains additional citation context not shown here]

Winograd, T. (1973). A procedural model of language understanding. In Schank, R. C. and Colby, K. M., editors, Computer models of thought and language, pages 152--186. W. H. Freeman.

....is another way of representing knowledge as a set of procedures for using knowledge. Procedural representation of a piece of information is a plan on how to use that information. In other words, knowledge is represented by rules about a particular problem. This approach was taken by Winograd [41, 42] in his SHRDLU system which takes English sentences as input and produces a set of procedures for doing what the statement requested. Each procedure has a set goals and there is a mechanism to satisfy this goal. In practice, very few systems use procedural knowledge as the only way to represent ....

T. Winograd. A procedural model of language understanding. In K.M. Colby and R.C. Schank, editors, Computer Models of Tought and Language. Freeman, 1973.

....methods are described in the following sections. The use of Pronouns The use of anaphoric references such as this; that or it allows a broad range of operations while minimizing the number of sentences to be recognized. Our system uses a primitive type of deduction to resolve these references [19]. After a sentence is recognized and executed we save the object, place or actor referenced by it. Then if in a later sentence there is an anaphoric reference it is resolved by using the most recent suitable saved reference. For instance when the user says Where is the newspaper, the object ....

Terry Winograd. A procedural model of language understanding. Computer models of thought and language. Schank & Colby. W.H? Freeman and Company San Francisco, 1973.

No context found.

Terry Winograd, A Procedural Model of Language Understanding, Computer Models of Thought and Language, R. Schank and K. Colby, eds., 152-186, New York: W.H. Freeman, 1973.

No context found.

Winograd, T: A procedural model of language understanding. In Schank, R., Colby, K. (eds.) : Computers models of thought and Language, Sand Francisco W.H. Freeman (1972)

No context found.

Terry Winograd. A procedural model of language understanding. In Computer Models of Thought and Language. W.H. Freeman and Company, 1973.

No context found.

Terry Winograd, "A Procedural Model of Language Understanding", Computer Models of Thought and Language, Roger Schank & Kenneth Colby eds., W. H. Freman Press, 1973.

No context found.

Terry Winograd. A procedural model of language understanding. In Readings in Natural Language Processing, pages 249--266. Morgan Kaufman, 1986.

No context found.

Terry Winograd, "A Procedural Model of Language Understanding", Computer Models of Thought and Language, Roger Schank & Kenneth Colby eds., W. H. Freman Press, 1973.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC