Pseudo-relaxation learning algorithm for complex-valued associative memory

HAM (Hopfield Associative Memory) and BAM (Bidirectinal Associative Memory) are representative associative memories by neural networks. The storage capacity by the Hebb rule, which is often used, is extremely low. In order to improve it, some learning methods, for example, pseudo-inverse matrix learning and gradient descent learning, have been introduced. Oh introduced pseudo-relaxation learning algorithm to HAM and BAM. In order to accelerate it, Hattori proposed quick learning. Noest proposed CAM (Complex-valued Associative Memory), which is complex-valued HAM. The storage capacity of CAM by the Hebb rule is also extremely low. Pseudo-inverse matrix learning and gradient descent learning have already been generalized to CAM. In this paper, we apply pseudo-relaxation learning algorithm to CAM in order to improve the capacity.     

An algorithm for linear metabolic pathway alignment

Metabolic pathway alignment represents one of the most powerful tools for comparative analysis of metabolism. It involves recognition of metabolites common to a set of functionally-related metabolic pathways, interpretation of biological evolution processes and determination of alternative metabolic pathways. Moreover, it is of assistance in function prediction and metabolism modeling. Although research on genomic sequence alignment is extensive, the problem of aligning metabolic pathways has received less attention. We are motivated to develop an algorithm of metabolic pathway alignment to reveal the similarities between metabolic pathways. A new definition of the metabolic pathway is introduced. The algorithm has been implemented into the PathAligner system; its web-based interface is available at http://bibiserv.techfak.uni-bielefeld.de/pathaligner/.     

BOSO: A novel feature selection algorithm for linear regression with high-dimensional data

With the frenetic growth of high-dimensional datasets in different biomedical domains, there is an urgent need to develop predictive methods able to deal with this complexity. Feature selection is a relevant strategy in machine learning to address this challenge. We introduce a novel feature selection algorithm for linear regression called BOSO (Bilevel Optimization Selector Operator). We conducted a benchmark of BOSO with key algorithms in the literature, finding a superior accuracy for feature selection in high-dimensional datasets. Proof-of-concept of BOSO for predicting drug sensitivity in cancer is presented. A detailed analysis is carried out for methotrexate, a well-studied drug targeting cancer metabolism.     

A genetic algorithm with adaptive mutations and family competition for training neural networks

In this paper, we present a new evolutionary technique to train three general neural networks. Based on family competition principles and adaptive rules, the proposed approach integrates decreasing-based mutations and self-adaptive mutations to collaborate with each other. Different mutations act as global and local strategies respectively to balance the trade-off between solution quality and convergence speed. Our algorithm is then applied to three different task domains: Boolean functions, regular language recognition, and artificial ant problems. Experimental results indicate that the proposed algorithm is very competitive with comparable evolutionary algorithms. We also discuss the search power of our proposed approach.     

An expectation maximization algorithm for training hidden substitution models

We derive an expectation maximization algorithm for maximum-likelihood training of substitution rate matrices from multiple sequence alignments. The algorithm can be used to train hidden substitution models, where the structural context of a residue is treated as a hidden variable that can evolve over time. We used the algorithm to train hidden substitution matrices on protein alignments in the Pfam database. Measuring the accuracy of multiple alignment algorithms with reference to BAliBASE (a database of structural reference alignments) our substitution matrices consistently outperform the PAM series, with the improvement steadily increasing as up to four hidden site classes are added. We discuss several applications of this algorithm in bioinformatics.     

A pilot study of combined working memory and inhibition training for children with AD/HD

Building on recent favourable outcomes using working memory (WM) training, this study examined the behavioural and physiological effect of concurrent computer-based WM and inhibition training for children with attention-deficit hyperactivity disorder (AD/HD). Using a double-blind active-control design, 29 children with AD/HD completed a 5-week at-home training programme and pre- and post-training sessions which included the assessment of overt behaviour, resting EEG, as well as task performance, skin conductance level and event-related potentials (ERPs) during a Go/Nogo task. Results indicated that after training, children from the high-intensity training condition showed reduced frequency of inattention and hyperactivity symptoms. Although there were trends for improved Go/Nogo performance, increased arousal and specific training effects for the inhibition-related N2 ERP component, they failed to reach standard levels of statistical significance. Both the low- and high-intensity conditions showed resting EEG changes (increased delta, reduced alpha and theta activity) and improved early attention alerting to Go and Nogo stimuli, as indicated by the N1 ERP component, post-training. Despite limitations, this preliminary work indicates the potential for cognitive training that concurrently targets the interrelated processes of WM and inhibition to be used as a treatment for AD/HD.     

A Monte Carlo EM algorithm for generalized linear mixed models with flexible random effects distribution

A popular way to represent clustered binary, count, or other data is via the generalized linear mixed model framework, which accommodates correlation through incorporation of random effects. A standard assumption is that the random effects follow a parametric family such as the normal distribution; however, this may be unrealistic or too restrictive to represent the data. We relax this assumption and require only that the distribution of random effects belong to a class of 'smooth' densities and approximate the density by the seminonparametric (SNP) approach of Gallant and Nychka (1987). This representation allows the density to be skewed, multi-modal, fat- or thin-tailed relative to the normal and includes the normal as a special case. Because an efficient algorithm to sample from an SNP density is available, we propose a Monte Carlo EM algorithm using a rejection sampling scheme to estimate the fixed parameters of the linear predictor, variance components and the SNP density. The approach is illustrated by application to a data set and via simulation.     

A constrained linear regression optimization algorithm for diaphragm motion tracking with cone beam CT projections

PurposeWe presented a feasibility study to extract the diaphragm motion from the inferior contrast cone beam computed tomography (CBCT) projection images using a constrained linear regression optimization algorithm.MethodsThe shape of the diaphragm was fitted by a parabolic function which was initialized by five manually placed points on the diaphragm contour of a pre-selected projection. A constrained linear regression model by exploiting the spatial, algebraic, and temporal constraints of the diaphragm, approximated by a parabola, was employed to estimate the parameters. The algorithm was assessed by a fluoroscopic movie acquired at anterior-posterior (AP) fixed direction and kilovoltage CBCT projection image sets from four lung and two liver patients using the Varian 21iX Clinac. The automatic tracing by the proposed algorithm and manual tracking were compared in both space and frequency domains for the algorithm evaluations.ResultsThe error between the results estimated by the proposed algorithm and those by manual tracking for the AP fluoroscopic movie was 0.54 mm with standard deviation (SD) of 0.45 mm. For the detected projections the average error was 0.79 mm with SD of 0.64 mm for all six enrolled patients and the maximum deviation was 2.5 mm. The mean sub-millimeter accuracy outcome exhibits the feasibility of the proposed constrained linear regression approach to track the diaphragm motion on rotational fluoroscopic images.ConclusionThe new algorithm will provide a potential solution to rendering diaphragm motion and possibly aiding the tumor target tracking in radiation therapy of thoracic/abdominal cancer patients.     

A novel algorithm with differential evolution and coral reef optimization for extreme learning machine training

Extreme learning machine (ELM) is a novel and fast learning method to train single layer feed-forward networks. However due to the demand for larger number of hidden neurons, the prediction speed of ELM is not fast enough. An evolutionary based ELM with differential evolution (DE) has been proposed to reduce the prediction time of original ELM. But it may still get stuck at local optima. In this paper, a novel algorithm hybridizing DE and metaheuristic coral reef optimization (CRO), which is called differential evolution coral reef optimization (DECRO), is proposed to balance the explorative power and exploitive power to reach better performance. The thought and the implement of DECRO algorithm are discussed in this article with detail. DE, CRO and DECRO are applied to ELM training respectively. Experimental results show that DECRO-ELM can reduce the prediction time of original ELM, and obtain better performance for training ELM than both DE and CRO.     

Recent advances in the MOBJ algorithm for training artificial neural networks

This paper presents a new scheme for training MLPs which employs a relaxation method for multi-objective optimization. The algorithm works by obtaining a reduced set of solutions, from which the one with the best generalization is selected. This approach allows balancing between the training error and norm of network weight vectors, which are the two objective functions of the multi-objective optimization problem. The method is applied to classification and regression problems and compared with Weight Decay (WD), Support Vector Machines (SVMs) and standard Backpropagation (BP). It is shown that the systematic procedure for training proposed results on good generalization neural models, and outperforms traditional methods.     

A cortical-hippocampal system for declarative memory

Recent neurobiological studies have begun to reveal the cognitive and neural coding mechanisms that underlie declarative memory--our ability to recollect everyday events and factual knowledge. These studies indicate that the critical circuitry involves bidirectional connections between the neocortex, the parahippocampal region and the hippocampus. Each of these areas makes a unique contribution to memory processing. Widespread high-order neocortical areas provide dedicated processors for perceptual, motor or cognitive information that is influenced by other components of the system. The parahippocampal region mediates convergence of this information and extends the persistence of neocortical memory representations. The hippocampus encodes the sequences of places and events that compose episodic memories, and links them together through their common elements. Here I describe how these mechanisms work together to create and re-create fully networked representations of previous experiences and knowledge about the world.     

A new training algorithm using artificial neural networks to classify gender-specific dynamic gait patterns

The aim of this study was to present a new training algorithm using artificial neural networks called multi-objective least absolute shrinkage and selection operator (MOBJ-LASSO) applied to the classification of dynamic gait patterns. The movement pattern is identified by 20 characteristics from the three components of the ground reaction force which are used as input information for the neural networks in gender-specific gait classification. The classification performance between MOBJ-LASSO (97.4%) and multi-objective algorithm (MOBJ) (97.1%) is similar, but the MOBJ-LASSO algorithm achieved more improved results than the MOBJ because it is able to eliminate the inputs and automatically select the parameters of the neural network. Thus, it is an effective tool for data mining using neural networks. From 20 inputs used for training, MOBJ-LASSO selected the first and second peaks of the vertical force and the force peak in the antero-posterior direction as the variables that classify the gait patterns of the different genders.     

A training primer for Institutional Officials

The laws and policies governing the care and use of animals in research in the US require institutions to establish training programs to assure that personnel are qualified for their roles in animal care and use programs. Few programs define specific training requirements for the Institutional Official (IO), one of the most important roles in an animal care program. In some cases, IOs may have little or no experience in biomedical science. In this article, the author provides an overview of the IO's role in an animal care and use program as defined by US government laws and policies for use in training IOs and chief executive officers. The author outlines the key responsibilities of the IO in an animal care program, the implications of noncompliance with federal requirements and some of the pitfalls in program design.     

Binary neural network training algorithms based on linear sequential learning

A key problem in Binary Neural Network learning is to decide bigger linear separable subsets. In this paper we prove some lemmas about linear separability. Based on these lemmas, we propose Multi-Core Learning (MCL) and Multi-Core Expand-and-Truncate Learning (MCETL) algorithms to construct Binary Neural Networks. We conclude that MCL and MCETL simplify the equations to compute weights and thresholds, and they result in the construction of simpler hidden layer. Examples are given to demonstrate these conclusions.     

A holographic model for associative memory chains

Holographic brain models are well suited to describe specific brain functions. Central nervous systems and holographic systems both show parallel information processing and non-localized storage in common. To process information both systems use correlation functions suggesting to develop cybernetical brain models in terms of holography. Associative holographic storage is done with two simultaneously existing patterns. They may reconstruct each other mutually. Time-sequentially existing patterns are connected to associative chains, if every two succeeding patterns do exist within a common period of time in order to be stored in pairs. Read out (recall) of associative chains—reconstructing coupled patterns which didn't exist simultaneously—requires advanced holographic techniques. Three different methods are described and tested experimentally. The underlying principles are feedback mechanisms, nonlinearities of the storage material and tridimensional architecture of the voluminous recording medium. Those principles evidently occur in neural storage systems supporting analogous information processing in neural- and holographic systems.