Hybrid neural network

About: Hybrid neural network is a research topic. Over the lifetime, 1305 publications have been published within this topic receiving 18223 citations.

Depth estimation from single monocular images using deep hybrid network

Abstract: Depth estimation is a significant task in the robotics vision. In this paper, we address the depth estimation from a single monocular image, which is a challenging problem in automated vision systems since a single image alone does not carry any additional measurements. To tackle our main objective, we design a deep hybrid neural network, which is composed of convolutional and recurrent layers (ReNet), where each ReNet layer is composed of the Long Short-Term Memory unit (LSTM), which is famous for the ability to memorize long-range context. In the proposed network, ReNet layers aim to enrich the features representation by directly capturing global context. The effective integration of ReNet and convolutional layers in the common CNN framework allows us to train the hybrid network in the end-to-end fashion. Experimental evaluation on the benchmarks dataset demonstrated, that hybrid network achieves the state-of-the-art results without any post-processing steps. Moreover, the composition of recurrent and convolutional layers provide more satisfying results.

A syntactic path-based hybrid neural network for negation scope detection

Abstract: The automatic detection of negation is a crucial task in a wide-range of natural language processing (NLP) applications, including medical data mining, relation extraction, question answering, and sentiment analysis. In this paper, we present a syntactic path-based hybrid neural network architecture, a novel approach to identify the scope of negation in a sentence. Our hybrid architecture has the particularity to capture salient information to determine whether a token is in the scope or not, without relying on any human intervention. This approach combines a bidirectional long short-term memory (Bi-LSTM) network and a convolutional neural network (CNN). The CNN model captures relevant syntactic features between the token and the cue within the shortest syntactic path in both constituency and dependency parse trees. The Bi-LSTM learns the context representation along the sentence in both forward and backward directions. We evaluate our model on the Bioscope corpus, and get 90.82% F-score (78.31% PCS) on the abstract sub-corpus, outperforming features-dependent approaches.

Hybrid neural network classifier, systems and methods

Abstract: A method of and system for parallelizing an program, comprising the steps of inputting an algorithm, operating said algorithm on selected data inputs to generate representative outputs, inputting representative outputs into parallelizing algorithms, and outputting a parallel implementation of said algorithm. In particular, this provides a parallel framework for target classification and pattern recognition procedures.

Hybrid-maximum neural network for depth analysis from stereo-image

Abstract: In present paper, we describe completely innovation architecture of artificial neural nets based on Hopfield structure for solving of stereo matching problem. Hybrid neural network consists of classical analogue Hopfield neural network and maximal neural network. The role of analogue Hopfield network is to find of attraction area of global minimum, whereas maximum network is to find accurate location of this minimum. Presented network characterizes by extremely high rate of working with the same accuracy as classical Hopfield-like network. It is very important as far as application and system of visually impaired people supporting are concerned. Considered network was taken under experimental tests with using real stereo pictures as well simulated stereo images. This allows on calculation of errors and direct comparison to classic analogue Hopfield neural network. Results of tests have shown, that the same accuracy of solution as for continuous Hopfield-like network, can be reached by described here structure in half number of classical Hopfield net iteration.