CRUR: coupled-recurrent unit for unification, conceptualization and context capture for language representation - a gene
- PDF / 8,083,289 Bytes
- 43 Pages / 439.642 x 666.49 pts Page_size
- 22 Downloads / 152 Views
CRUR: coupled-recurrent unit for unification, conceptualization and context capture for language representation - a generalization of bi directional LSTM Chiranjib Sur1 Received: 21 July 2019 / Revised: 6 September 2020 / Accepted: 9 September 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this work we have analyzed a novel concept of sequential binding based learning capable network based on the coupling of recurrent units with Bayesian Prior definition. The coupling structure encodes to generate efficient tensor representations that can be decoded to generate efficient sentences and can describe certain events. These descriptions are derived from structural representations of visual features of images and media. An elaborated study of the different types of coupling recurrent structures are studied and some insights of their performance are provided. Supervised learning performance for natural language processing is judged based on statistical evaluations, however, the truth is perspective, and in this case the qualitative evaluations reveal the real capability of the different architectural strengths and variations. Bayesian Prior definition of different embedding helps in better characterization of the sentences based on the natural language structure related to parts of speech and other semantic level categorization in a form which is machine interpret-able and inherits the characteristics of the Tensor Representation binding and unbinding based on the mutually orthogonality. Our approach has surpassed some of the existing basic works related to image captioning. Keywords Language modeling · Dual context initialization · Representation learning · Tensor representation · Memory networks
1 Introduction Long short term (LSTM) memories are widely analyzed due to their high demand in industry to tackle huge volume of unlabeled data, and data analytic technologies greatly rely on them. Mere object detection and manual tagging failed to provide immense details of the activities and the events in the media data and to overcome the confusion created due Chiranjib Sur
[email protected] 1
Computer & Information Science & Engineering Department, University of Florida, Gainesville, FL, USA
Multimedia Tools and Applications
to perception and language barriers between human interpretation capability and machine interpretation. Image captioning [49] has progressed but slowed down to gain the optimum efficiency and in this work we have analyzed a new architecture that enhances the image captioning problem from visual features. In disguise, we introduced an effective way of coupling and decoupling tensors which can gather effective representations that can differentiate between different ways of writing and sentence constructions. The new architecture, named Coupled-Recurrent Unit Representation (CRUR) unit, is based on the entanglement of the representation of two recurrent units and passing the knowledge into a form of a structured Tensor Product Representations and decoup
Data Loading...
