Koprinkova-Hristova P., Mladenov V., Kasabov N.K. (eds.) Artificial Neural Networks. Methods and Applications in Bio-/Neuroinformatics

Springer, 2015. — 487 p.The edited book includes chapters that present selected and extended papers from the International Conference on Artificial Neural Networks (ICANN) 2013. Founded in 1991, the ICANN become the Premier annual conference of the European Neural Network Society. Its main goal is to bring together and to facilitate contacts between researchers from information sciences and neurosciences and to provide a high-level international forum for both academic and industrial communities.
The selected and invited to the present book chapters were selected among ICANN papers that received highest scores during the strong peer review assessment (almost 40% rejection rate) and among selected by session chairs best presentations.
The collected in the book chapters are presented in topical sections that cover wide range of contemporary topics varying from neural network theory and models, machine learning and learning algorithms, brain-machine interaction and bio-inspired systems, pattern recognition and classification as well as various applications.
The book chapters include new theoretical developments in recurrent neural networks and reservoir computing, new and improved training algorithms for Deep Boltzmann Machines (DBM), tapped delay feedforward architectures and kernel machines, reinforcement learning and Adaptive Critic Designs (ACD), new bioinspired models and architectures related to cell assembly mechanisms, visual perception and natural language understanding, new and improved algorithms for pattern recognition with applications to gesture classification, handwritten digit recognition and time series forecasting.
The book will be of interest to all researchers and postgraduate students in the area of computational intelligence, applied mathematics, computer science, engineering, neuroscience, and other related areas.Neural Networks Theory and Models
Recurrent Neural Networks and Super-Turing Interactive Computation
Image Classification with Nonnegative Matrix Factorization Based on Spectral Projected Gradient
Energy-Time Tradeoff in Recurrent Neural Nets
An Introduction to Delay-Coupled Reservoir Computing
Double-Layer Vector Perceptron for Binary Patterns Recognition
Local Detection of Communities by Attractor Neural-Network Dynamics
Learning Gestalt Formations for Oscillator Networks
Analysing the Multiple Timescale Recurrent Neural Network for Embodied Language Understanding
Learning to Look and Looking to Remember: A Neural-Dynamic Embodied Model for Generation of Saccadic Gaze Shifts and Memory Formation
New Machine Learning Algorithms for Neural Networks
How to Pretrain Deep Boltzmann Machines in Two Stages
Training Dynamic Neural Networks Using the Extended Kalman Filter for Multi-Step-Ahead Predictions
Learning as Constraint Reactions
Baseline-Free Sampling in Parameter Exploring Policy Gradients: Super Symmetric PGPE
Sparse Approximations to Value Functions in Reinforcement Learning
Neural Networks Solution of Optimal Control Problems with Discrete Time Delays and Time-Dependent Learning of Infinitesimal Dynamic System
Pattern Recognition, Classification and Other Neural Network Applications
Applying Prototype Selection and Abstraction Algorithms for Efficient Time-Series Classification
Enforcing Group Structure through the Group Fused Lasso
Incremental Anomaly Identification in Flight Data Analysis by Adapted One-Class SVM
Inertial Gesture Recognition with BLSTM-RNN
Online Recognition of Fixations, Saccades, and Smooth Pursuits for Automated Analysis of Traffic Hazard Perception
Input Transformation and Output Combination for Improved Handwritten Digit Recognition
Feature Selection for Interval Forecasting of Electricity Demand Time Series Data
Stacked Denoising Auto-Encoders for Short-Term Time Series Forecasting