Large number of weights in deep neural networks makes the models difficult to be deployed in low memory environments such as, mobile phones, IOT edge devices as well as “inferencing as a service” environments on cloud. Prior work has considered reduction in the size of the models, through compression techniques like pruning, quantization, Huffman encoding etc. However, efficient inferencing usi...

PURPOSE A hallmark of autistic spectrum disorder (ASD) is disruption in theory-of-mind development, including the understanding of false beliefs. Previous studies have typically assessed the development of first- and second-order false belief concepts in ASD, with tasks primarily emphasizing physical causality and logical inferencing. The present study investigated how preadolescents and adoles...

PURPOSE Many children with communication disorders have reading comprehension difficulties, and in order to target interventions effectively it is important to identify which specific components of comprehension are especially challenging. The current study explored the relationship between text-inferencing skill, autistic symptomatology, and language phenotype. METHOD Typically developing ch...

We have been addressing the problem of providing access to the free text in biomedical databases. The focus of our work is the development of SPECIALIST, an experimental NLP system for the biomedical domain. The system includes a broad coverage parser supported by a large lexicon, a module that provides acces s to extensive biomedical knowledge sources, and a retrieval module that allows us to ...

The power of belief networks lies in its connective edges where the influences are bidirectional. While Bayesian methods capture bidirectional influences, we propose a simpler and faster method of inferencing from nodal observations that uses bidirectional fuzzy influences that are propagated via fuzzy set membership functions. We need neither the conditional probability tables nor constraining...

Dialogue inferencing is the knowledge-intensive process of inferring aspects of a user's problem from its partial description. Conversational case-based reasoning (CCBR) systems, which interactively and incrementally elicit a user's problem description , suuer from poor retrieval eeciency (i.e., they prompt the user with questions that the user has already implicitly answered) unless they perfo...

As a computational model of natural language communication, Database Semantics1 (DBS) includes a hearer mode and a speaker mode. For the content to be mapped into language expressions, the speaker mode requires an autonomous control. The control is driven by the overall task of maintaining the agent in a state of balance by connecting the interfaces for recognition with those for action. This p...

Project Halo has the long-term objective of developing a digital Aristotle, i.e. a knowledge system that is able to answer questions in a particular domain and give explanations for its answers. In this paper we report about the Ontoprise contribution to the Halo Pilot Project, in which various competing ontology engineering methodologies and knowledge system capabilities have been investigated...