Considering the difficulties inherent in the manual construction of natural language parsers, we have designed and implemented our system GRIND which is capable of learning a sequence of context-dependent parsing actions from an arbitrary corpus containing labelled parse trees. Being trained and tested on corpus SUSANNE, GRIND reaches the accuracy of 96 % and the recall of 68 %.

A classic result of Johnson and Lindenstrauss asserts that any set of n points in d-dimensional Euclidean space can be embedded into k-dimensional Euclidean space—where k is logarithmic in n and independent of d—so that all pairwise distances are maintained within an arbitrarily small factor. All known constructions of such embeddings involve projecting the n points onto a spherically random k-...

This paper argues first that, contrary to what one would expect, metaphorical interpretations of utterances pass two of Cappelan and Lepore’s Minimalist tests for semantic context-sensitivity. I then propose how, in light of that result, one might analyze metaphors on the model of indexicals and demonstratives, expressions that (even) Minimalists agree are semantically context-dependent. This a...

This paper describes our eeorts in designing a two-stage recognizer with the objective of developing a multi-domain speech understanding system. We envisage one rst-stage recognition engine that is domain-independent, and multiple second-stage systems specializing in individual domains. A major novelty in our initial two-stage design is a front-end that incorporates angie-based hierarchical sub...

The relative gain array (RGA) and condition number are commonly used tools in controllability analysis. In this paper we present new results that link these measures to control performance, measured in terms of the output sensitivity function with input and output uncertainty.

Retrieval dynamics in context-dependent recall were explored via manipulations of external and internal context in two experiments. Participants were tested in either the same or different context as the material was learned in and correct recalls, errors, and recall latency measures were examined. In both experiments changes in context resulted in fewer correct items being recalled than when c...

In β, contrary to what was the case in α, sentence (2) appears to be false. Examples like the card scenario can be multiplied, and seem to lead to the conclusion that the truth of sentences where ‘knowing which’ or ‘knowing who’ (henceforth ‘knowing-wh’) constructions occur is context-dependent. Consider again contexts α and β. It is appealing to adopt the view that one of the roles that these ...

Several errors in Refs. [1, 2] are corrected related to the optical trapping potentials for a stateinsensitive, compensated nanofiber trap for the D2 transition of atomic Cesium. Section I corrects our basic formalism in Ref. [1] for calculating dipole-force potentials. Section II corrects erroneous values for a partial lifetime and a transition wavelength in Ref. [1]. Sections III and IV prese...

We propose the foundations of a computer model of scientific discovery that takes into account certain psychological aspects of human observation of the world. To this end, we simulate two main components of such a system. The first is a dynamic microworld in which physical events take place, and the second is an observer that visually perceives entities and events in the microworld. For reason...

What are the best ways for design teams attend to issues of power, inequity, trust and other ethical concerns as they arise in design? Literature on value-sensitive design (VSD) and technology ethics has advocated for a range of design methods that propose different roles and responsibility for ethics during technology development. This paper explores four provocations that imagine different ro...