Developing structured representations

Developing Distributed Representations of structured sequences by Autoassociation

1 Introduction In spite of the growing research on connectionist Natural Language Processing (NLP), there are still a number of challenges to be solved, for example, the development of proper linguistic representations. Natural language is a dynamic system with underlying hierarchical structure and sequential external appearance. Therefore, NLP systems need an adequate hierarchical system of li...

Learning structured representations

shruti is a connectionist model that demonstrates how a network of neuron-like elements can encode a large body of semantic, episodic, and causal knowledge, and rapidly make decisions and perform explanatory and predictive reasoning. To further ground this model in the functioning of the brain it must be shown that components of the model can be learned in a neurally plausible manner. Previous ...

Learning Structured Text Representations

In this paper, we focus on learning structureaware document representations from data without recourse to a discourse parser or additional annotations. Drawing inspiration from recent efforts to empower neural networks with a structural bias (Cheng et al., 2016; Kim et al., 2017), we propose a model that can encode a document while automatically inducing rich structural dependencies. Specifical...

Structured Representations of Empirical Information

We assessed in a pilot study the effect on some aspects of renal flmction of 6 weeks’ administration of a combination of aspirin-dipyridamole (990 mg/225 mg daily) administered on a double-blind crossover schedule in 16 insulin-dependent diabetic patients with nephropathy. Total 24-h urinary protein excretion (16 patients) was significantly reduced during aspirin-dipyridamole administration fro...

Recurrent Autoassociative Networks: Developing Distributed Representations of Hierarchically Structured Sequences by Autoassociation