Detection of synchronized firings in multivariate neural spike trains during motor tasks.

This paper describes and compares two classical methods for the detection of neuron groups which exhibit synchronized firings in multivariate spike trains. These methods were compared on experimental and randomized data corresponding to the firing activity of 104 neurons located in motor, premotor, and parietal cortices in a monkey during movement tasks. Both methods exhibited high false positive rates in randomized data, but results showed that this rate can be advantageously reduced with a simple postprocessing. Otherwise, one method permitted to detect a significant number of synchronized groups of neurons related to the behavioral task.

A comparison of optimal MIMO linear and nonlinear models for brain-machine interfaces.

The field of brain-machine interfaces requires the estimation of a mapping from spike trains collected in motor cortex areas to the hand kinematics of the behaving animal. This paper presents a systematic investigation of several linear (Wiener filter, LMS adaptive filters, gamma filter, subspace Wiener filters) and nonlinear models (time-delay neural network and local linear switching models) applied to datasets from two experiments in monkeys performing motor tasks (reaching for food and target hitting). Ensembles of 100-200 cortical neurons were simultaneously recorded in these experiments, and even larger neuronal samples are anticipated in the future. Due to the large size of the models (thousands of parameters), the major issue studied was the generalization performance. Every parameter of the models (not only the weights) was selected optimally using signal processing and machine learning techniques. The models were also compared statistically with respect to the Wiener filter as the baseline. Each of the optimization procedures produced improvements over that baseline for either one of the two datasets or both.

Visualization of the short term maximum Lyapunov exponent topography in the epileptic brain.

In this paper, a new kind of brain topography is introduced and applied to data from four patients affected by intractable epilepsy. Experience has shown that the short term maximum Lyapunov exponent (STLmax) is a robust parameter when optimized for the dynamical analysis of the electroencephalography (EEG). The objective of this work is to map the spatial distribution of STLmax over time. STLmax is estimated from segments of each channel of long term continuous scalp EEG recordings and a movie of the STLmax segment estimates is created over the head. Movies allow for a simple visualization of which electrodes are related to the highest or lowest chaoticity for the longest time. We found out that the interictal epileptiform activity is related to the highest STLmax level, whereas the focal area is related to low STLmax levels during either the interictal and preictal stages.

Simultaneus prediction of four kinematic variables for a brain-machine interface using a single recurrent neural network.

Implementation of brain-machine interface neural-to-motor mapping algorithms in low-power, portable digital signal processors (DSPs) requires efficient use of model resources especially when predicting signals that show interdependencies. We show here that a single recurrent neural network can simultaneously predict hand position and velocity from the same ensemble of cells using a minimalist topology. Analysis of the trained topology showed that the model learns to concurrently represent multiple kinematic parameters in a single state variable. We further assess the expressive power of the state variables for both large and small topologies.

Generalized information potential criterion for adaptive system training.

We have previously proposed the quadratic Renyi's error entropy as an alternative cost function for supervised adaptive system training. An entropy criterion instructs the minimization of the average information content of the error signal rather than merely trying to minimize its energy. In this paper, we propose a generalization of the error entropy criterion that enables the use of any order of Renyi's entropy and any suitable kernel function in density estimation. It is shown that the proposed entropy estimator preserves the global minimum of actual entropy. The equivalence between global optimization by convolution smoothing and the convolution by the kernel in Parzen windowing is also discussed. Simulation results are presented for time-series prediction and classification where experimental demonstration of all the theoretical concepts is presented.

Extraction of principal components from biosignals by neural net.

This contribution gives the information on a useful application of principal component analysis (PCA) in the field of electroencephalogram (EEG) and laser-Doppler signal processing. The principal components are estimated by a neural network (NN) approach.

Super-resolution of images based on local correlations.

An adaptive two-step paradigm for the superresolution of optical images is developed in this paper. The procedure locally projects image samples onto a family of kernels that are learned from image data. First, an unsupervised feature extraction is performed on local neighborhood information from a training image. These features are then used to cluster the neighborhoods into disjoint sets for which an optimal mapping relating homologous neighborhoods across scales can be learned in a supervised manner. A super-resolved image is obtained through the convolution of a low-resolution test image with the established family of kernels. Results demonstrate the effectiveness of the approach.

Training neural networks with additive noise in the desired signal.

A new global optimization strategy for training adaptive systems such as neural networks and adaptive filters [finite or infinite impulse response (FIR or IIR)] is proposed in this paper. Instead of adding random noise to the weights as proposed in the past, additive random noise is injected directly into the desired signal. Experimental results show that this procedure also speeds up greatly the backpropagation algorithm. The method is very easy to implement in practice, preserving the backpropagation algorithm and requiring a single random generator with a monotonically decreasing step size per output channel. Hence, this is an ideal strategy to speed up supervised learning, and avoid local minima entrapment when the noise variance is appropriately scheduled.

Target discrimination in synthetic aperture radar using artificial neural networks.

This paper addresses target discrimination in synthetic aperture radar (SAR) imagery using linear and nonlinear adaptive networks. Neural networks are extensively used for pattern classification but here the goal is discrimination. We show that the two applications require different cost functions. We start by analyzing with a pattern recognition perspective the two-parameter constant false alarm rate (CFAR) detector which is widely utilized as a target detector in SAR. Then we generalize its principle to construct the quadratic gamma discriminator (QGD), a nonparametrically trained classifier based on local image intensity. The linear processing element of the QCD is further extended with nonlinearities yielding a multilayer perceptron (MLP) which we call the NL-QGD (nonlinear QGD). MLPs are normally trained based on the L(2) norm. We experimentally show that the L(2) norm is not recommended to train MLPs for discriminating targets in SAR. Inspired by the Neyman-Pearson criterion, we create a cost function based on a mixed norm to weight the false alarms and the missed detections differently. Mixed norms can easily be incorporated into the backpropagation algorithm, and lead to better performance. Several other norms (L(8), cross-entropy) are applied to train the NL-QGD and all outperformed the L(2) norm when validated by receiver operating characteristics (ROC) curves. The data sets are constructed from TABILS 24 ISAR targets embedded in 7 km(2) of SAR imagery (MIT/LL mission 90).

Incremental backpropagation learning networks.

How to learn new knowledge without forgetting old knowledge is a key issue in designing an incremental-learning neural network. In this paper, we present a new incremental learning method for pattern recognition, called the "incremental backpropagation learning network", which employs bounded weight modification and structural adaptation learning rules and applies initial knowledge to constrain the learning process. The viability of this approach is demonstrated for classification problems including the iris and the promoter domains.