The paper first summarizes a general approach to the training of recurrent neural networks by gradient-based algorithms, which leads to the introduction of four families of training algorithms. Because of the variety of possibilities thus available to the "neural network designer", the choice of the appropriate algorithm to solve a given problem becomes critical. We show that, in the case of pr...

Building practical and efficient intrusion detection systems in computer network is important in industrial areas today and machine learning technique provides a set of effective algorithms to detect network intrusion. To find out appropriate algorithms for building such kinds of systems, it is necessary to evaluate various types of machine learning algorithms based on specific criteria. In thi...

Alpenglow1 is a free and open source C++ framework with easy-touse Python API. Alpenglow is capable of training and evaluating industry standard recommendation algorithms including variants of popularity, nearest neighbor, and factorization models. Traditional recommender algorithms may periodically rebuild their models, but they cannot adjust online to quick changes in trends. Besides batch tr...

This paper presents a collapsed variational Bayesian inference algorithm for PCFGs that has the advantages of two dominant Bayesian training algorithms for PCFGs, namely variational Bayesian inference and Markov chain Monte Carlo. In three kinds of experiments, we illustrate that our algorithm achieves close performance to the Hastings sampling algorithm while using an order of magnitude less t...

Gaze and face tracking algorithms have traditionally battled a compromise between computational complexity and accuracy; the most accurate neural net algorithms cannot be implemented in real time, but less complex realtime algorithms suffer from higher error. This project seeks to better bridge that gap by improving on real-time eye and facial recognition algorithms in order to develop accurate...

Substantial number of problems in artificial intelligence requires optimization. Increasing complexity of the problems imposes several challenges on optimization algorithms. The algorithms must be fast, computationally efficient, and scalable. Balance between convergence speed and computational complexity is of central importance. Typical example is the task of training neural networks. Superli...

The training algorithm of Wavelet Neural Networks (WNN) is a bottleneck which impacts on the accuracy of the final WNN model. Several methods have been proposed for training the WNNs. From the perspective of our research, most of these algorithms are iterative and need to adjust all the parameters of WNN. This paper proposes a one-step learning method which changes the weights between hidden la...

Various algorithms for constructing weight matrices for Hopfield-type associative memories are reviewed, including ones with much higher capacity than the basic model. These alternative algorithms either iteratively approximate the projection weight matrix or use simple perceptron learning. An experimental investigation of the performance of networks trained by these algorithms is presented, in...

The development of engineering applications of neural networks makes it necessary to clarify the similarities and differences between the concepts and methods developed for neural networks and those used in more classical fields such as filtering and control. In previous papers [Nerrand et al. 1993], [Marcos et al. 1993], the relationships between non-linear adaptive filters and neural networks...

We describe new algorithms for training tagging models, as an alternative to maximum-entropy models or conditional random fields (CRFs). The algorithms rely on Viterbi decoding of training examples, combined with simple additive updates. We describe theory justifying the algorithms through a modification of the proof of convergence of the perceptron algorithm for classification problems. We giv...