Showing papers by "Charles W. Anderson published in 2011"

Critical issues in state-of-the-art brain–computer interface signal processing

Abstract: This paper reviews several critical issues facing signal processing for brain–computer interfaces (BCIs) and suggests several recent approaches that should be further examined. The topics were selected based on discussions held during the 4th International BCI Meeting at a workshop organized to review and evaluate the current state of, and issues relevant to, feature extraction and translation of field potentials for BCIs. The topics presented in this paper include the relationship between electroencephalography and electrocorticography, novel features for performance prediction, time-embedded signal representations, phase information, signal non-stationarity, and unsupervised adaptation.

Classification of EEG during imagined mental tasks by forecasting with Elman Recurrent Neural Networks

Abstract: The ability to classify EEG recorded while a subject performs varying imagined mental tasks may lay the foundation for building usable Brain-Computer Interfaces as well as improve the performance of EEG analysis software used in clinical settings. Although a number of research groups have produced EEG classifiers, these methods have not yet reached a level of performance that is acceptable for use in many practical applications. We assert that current approaches are limited by their ability to capture the temporal and spatial patterns contained within EEG. In order to address these problems, we propose a new generative technique for EEG classification that uses Elman Recurrent Neural Networks. EEG recorded while a subject performs one of several imagined mental tasks is first modeled by training a network to forecast the signal a single step ahead in time. We show that these models are able to forecast EEG well with an RMSE as low as 0.110. A separate model is then trained over EEG belonging to each class. Classification of previously unseen data is performed by applying each model and assigning the class label associated with the network that produced the lowest forecasting error. This approach is tested on EEG collected from two able-bodied subjects and one subject with a high-level spinal cord injury. Classification rates as high as 93.3% are achieved for a two-task problem with decisions made every second yielding a bitrate of 38.7 bits per minute.

Reliable identification of mental tasks using time-embedded EEG and sequential evidence accumulation

Abstract: Eleven channels of EEG were recorded from a subject performing four mental tasks. A time-embedded representation of the untransformed EEG samples was constructed. Classification of the time-embedded samples was performed by linear and quadratic discriminant analysis and by an artificial neural network. A classifier's output for consecutive samples is combined to increase reliability. A new performance measure is defined as the number of correct selections that would be made by a brain–computer interface (BCI) user of the system, accounting for the need for an incorrect selection to be followed by a correct one to 'delete' the previous selection. A best result of 0.32 correct selections s−1 (about 3 s per BCI decision) was obtained with a neural network using a time-embedding dimension of 50.

Stable reinforcement learning with recurrent neural networks

Abstract: In this paper, we present a technique for ensuring the stability of a large class of adaptively controlled systems. We combine IQC models of both the controlled system and the controller with a method of filtering control parameter updates to ensure stable behavior of the controlled system under adaptation of the controller. We present a specific application to a system that uses recurrent neural networks adapted via reinforcement learning techniques. The work presented extends earlier works on stable reinforcement learning with neural networks. Specifically, we apply an improved IQC analysis for RNNs with time-varying weights and evaluate the approach on more complex control system.

Comparison of EEG blind source separation techniques to improve the classification of P300 trials

Abstract: This paper provides a comparison of several blind source separation (BSS) techniques as they are applied to EEG signals. Specifically, this work focuses on the P300 speller paradigm and assesses the classification accuracies for the identification of P300 trials. Previous work has shown that BSS methods such as independent component analysis (ICA) are useful in extracting the P300 source information from the background noise, increasing the classification rates. ICA will be compared with two other BSS methods, maximum noise fraction (MNF) and principal component analysis (PCA). In addition to this, we will analyze the effect of adding temporal information to the original data, which allows these BSS algorithms to find more complex spatio-temporal patterns.

Discount and speed/execution tradeoffs in Markov Decision Process games

Abstract: We study Markov Decision Process (MDP) games with the usual ±1 reinforcement signal. We consider the scenario in which the goal of the game, rather than just winning, is to maximize the number of wins in an allotted period of time (or maximize the expected reward in the same period). In the reinforcement learning literature, this type of tradeoff is often handled by tuning the discount parameter in order to encourage the learning algorithm to find policies that take fewer steps on average, at the cost of a lower probability of winning. We show that this approach is not guaranteed to solve the tradeoff problem optimally, and hence a different strategy is needed when tackling this type of problems.

Hurricane Prediction with Python

Abstract: The National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) is a global spectral model used for aviation weather forecast. It produces forecasts of wind speed and direction, temperature, humidity and precipitation out to 192 hr every 6 hours over the entire globe. The horizontal resolution in operational version of the GFS is about 25 km. Much longer integration of similar global models are run for climate applications but with much lower horizontal resolution. Although not specifically designed for tropical cyclones, the model solutions contain smoothed representations of these storms. One of the challenges in using global atmospheric model for hurricane applications is objectively determining what is a tropical cyclone, given the three dimensional solutions of atmospheric variables. This is especially difficult in the lower resolution climate models. To address this issue, without manually selecting features of interests, the initial conditions from a low resolution version of the GFS (2 degree latitude-longitude grid) are examined at 6 hour periods and compared with the known positions of tropical cyclones. Several Python modules are used to build a prototype model quickly, and the prototype model shows fast and accurate prediction with the low resolution GFS data.