Hierarchical Reasoning with Distributed Vector Representations
Abstract
We demonstrate that distributed vector representations are capable of hierarchical reasoning by summing sets of vectors representing hyponyms (subordinate concepts) to yield a vector that resembles the associated hypernym (superordinate concept). These distributed vector representations constitute a potentially neurally plausible model while demonstrating a high level of performance in many different cognitive tasks. Experiments were run using DVRS, a word embedding system designed for the Sigma cognitive architecture, and Word2Vec, a state-of-the-art word embedding system. These results contribute to a growing body of work demonstrating the various tasks on which distributed vector representations perform competently.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2015
- Accession Number
- AD1158645
Entities
People
- Abram Demski
- Cody Kommers
- Paul Rosenbloom
- Volkan Ustun
Organizations
- University of Southern California