This paper describes a neural network model that directs saccades back to targets after they disappear and other saccades intervene. This is a simple example of knowing where something is after it is no longer visible and the observer has moved. These tasks require a short-term memory that can store continuous values of spatial location. The model was generated by training a neural network with...

We report on the application of recurrent neural nets in a openset text-dependent speaker identification task. The motivation for applying recurrent neural nets to this domain is to find out if their ability to take short-term spectral features but yet respond to long-term temporal events is advantageous for speaker identification. We use a feedforward net architecture adapted from that introdu...

Markov networks (MNs) can incorporate arbitrarily complex features in modeling relational data. However, this flexibility comes at a sharp price of training an exponentially complex model. To address this challenge, we propose a novel relational learning approach, which consists of a restricted class of relational MNs (RMNs) called relation tree-based RMN (treeRMN), and an efficient Hidden Vari...

MOTIVATION Dynamic programming is the core algorithm of sequence comparison, alignment and linear hidden Markov model (HMM) training. For a pair of sequence lengths m and n, the problem can be solved readily in O(mn)time and O(mn)space. The checkpoint algorithm introduced by Grice et al. (CABIOS, 13, 45--53, 1997) runs in O(Lmn)time and O(Lm(L) square root of n)space, where L is a positive inte...

In the last years there has been increasing interest in developing discriminative training methods for hidden Markov models, with the aim to improve their performance in classification and pattern recognition tasks. Although several advances have been made in this area, they have been targeted almost exclusively to standard models whose conditional observations are given by a Gaussian mixture d...

This paper concerns open-world classification, where the classifier not only needs to classify test examples into seen classes that have appeared in training but also reject examples from unseen or novel classes that have not appeared in training. Specifically, this paper focuses on discovering the hidden unseen classes of the rejected examples. Clearly, without prior knowledge this is difficul...

This paper presents performance analysis of the Evolutionary Programming-Artificial Neural Network (EPANN) based technique to optimize the architecture and training parameters of a one-hidden layer feedforward ANN model for the prediction of energy output from a grid connected photovoltaic system. The ANN utilizes solar radiation and ambient temperature as its inputs while the output is the tot...

Unsupervised neural network learning extracts hidden features from unlabeled training data. This is used as a pretraining step for further supervised learning in deep networks. Hence, understanding unsupervised learning is of fundamental importance. Here, we study the unsupervised learning from a finite number of data, based on the restricted Boltzmann machine where only one hidden neuron is co...

We present a hybrid architecture of recurrent neural networks (RNNs) inspired by hidden Markov models (HMMs). We train the hybrid architecture using genetic algorithms to learn and represent dynamical systems. We train the hybrid architecture on a set of deterministic finite-state automata strings and observe the generalization performance of the hybrid architecture when presented with a new se...

Conventional speech recognition systems are based on Gaussian hidden Markov models (HMMs). Discriminative techniques such as log-linear modeling have been investigated in speech recognition only recently. This thesis establishes a log-linear modeling framework in the context of discriminative training criteria, with examples from continuous speech recognition, part-of-speech tagging, and handwr...