We explore di‚erent approaches to integrating a simple convolutional neural network (CNN) with the Lucene search engine in a multi-stage ranking architecture. Our models are trained using the PyTorch deep learning toolkit, which is implemented in C/C++ with a Python frontend. One obvious integration strategy is to expose the neural network directly as a service. For this, we use Apache Œri‰, a ...

A connection is drawn between rational functions, the realization theory of dynamical systems, and feedforward neural networks. This allows us to parametrize single hidden layer scalar neural networks with (almost) arbitrary analytic activation functions in terms of strictly proper rational functions. Hence, we can solve the uniqueness of parametrization problem for such networks.

Augmenting responses in neocortical pyramidal cells can be elicited by cortical or thalamic repetitive stimulation around 10Hz. A realistic model of a cortical pyramidal (PY) cell and an interneuron (IN) was developed to explore possible intracortical mechanisms. The interaction between strong feedforward hyperpolarizing inhibition, deinactivation of a low-threshold Ca2` current and depression ...

This article proposes a modularization scheme for feedforward networks based on controllable internal representations. Control is achieved by replacing hidden units with pretrained modules that constrain internal patterns of activity to desired subsets. In the case of auto-associative feedforward networks these subsets can be seen as module interfaces. If enough a priori knowledge about a syste...

Absfracf-This paper presents a neural-tuned neural nehvork, which is trained by genetic algorithm (CA). The neural-tuned neural nehvork consists of a neural network and a modified neural network. I n the modified neural network, a neuron model with hvo activation functions is introduced. Some parameters of these activation functions will be tuned by neural network. The proposed network structur...

A feedforward neural network model is presented in this study to predict the execution time of a parallel Monte-Carlo implementation. The enormous performance range offered by today’s systems caused the performance evaluation tools to become more complicated to be able to consider the relative values and interrelated parameters. Artificial Neural Networks provide an excellent alternative to con...