Regularization has been one of the most popular approaches to prevent overfitting in electroencephalogram (EEG) classification of brain-computer interfaces (BCIs). The effectiveness of regularization is often highly d...
详细信息
Regularization has been one of the most popular approaches to prevent overfitting in electroencephalogram (EEG) classification of brain-computer interfaces (BCIs). The effectiveness of regularization is often highly dependent on the selection of regularization parameters that are typically determined by cross-validation (CV). However, the CV imposes two main limitations on BCIs: 1) a large amount of training data is required from the user and 2) it takes a relatively long time to calibrate the classifier. These limitations substantially deteriorate the system's practicability and may cause a user to be reluctant to use BCIs. In this paper, we introduce a sparsebayesian method by exploiting Laplace priors, namely, SBLaplace, for EEG classification. A sparse discriminant vector is learned with a Laplace prior in a hierarchical fashion under a bayesian evidence framework. All required model parameters are automatically estimated from training data without the need of CV. Extensive comparisons are carried out between the SBLaplace algorithm and several other competing methods based on two EEG data sets. The experimental results demonstrate that the SBLaplace algorithm achieves better overall performance than the competing algorithms for EEG classification.
Probabilistic wind generation forecast results are crucial for power system operational dispatch. In this paper, a nonparametric approach for short-term probabilistic wind generation forecast based on the sparse Bayes...
详细信息
ISBN:
(纸本)9781479983933
Probabilistic wind generation forecast results are crucial for power system operational dispatch. In this paper, a nonparametric approach for short-term probabilistic wind generation forecast based on the sparse bayesian classification (SBC) and Dempster-Shafer theory (DST) is proposed. This approach is composed by the following four steps: Firstly, a spot forecast of wind generation is performed based on Support Vector Machine (SVM);Secondly, the range of SVM forecast error is discretized into multiple intervals, and the conditional probability of each interval is estimated by a sparsebayesian classifier;Thirdly, DST is applied to combine the probabilities of all the intervals to form a unified probability distribution function of the SVM forecast error;Lastly, the probability distribution function of wind generation is achieved by combining the SVM wind generation spot forecast result and corresponding forecast error distribution. The distinguishing features of the proposed approach are as follows: (a) The approach is a nonparametric one and the forecast error caused by the misjudgement of probability distribution type can be avoided;(b) The proposed approach has good generalization capability by using the sparse learning mechanism;and (c) The range constraint of wind generation can be systematically considered in the approach by applying DST. Tests on a 74-MW wind farm illustrate the effectiveness of the proposed approach.
This article focuses on online occupancy mapping and real-time collision checking onboard an autonomous robot navigating in a large unknown environment. Commonly used voxel and octree map representations can be easily...
详细信息
This article focuses on online occupancy mapping and real-time collision checking onboard an autonomous robot navigating in a large unknown environment. Commonly used voxel and octree map representations can be easily maintained in a small environment but have increasing memory requirements as the environment grows. We propose a fundamentally different approach for occupancy mapping, in which the boundary between occupied and free space is viewed as the decision boundary of a machine learning classifier. This work generalizes a kernel perceptron model which maintains a very sparse set of support vectors to represent the environment boundaries efficiently. We develop a probabilistic formulation based on relevance vector machines, handling measurement noise, and probabilistic occupancy classification, supporting autonomous navigation. We provide an online training algorithm, updating the sparsebayesian map incrementally from streaming range data, and an efficient collision-checking method for general curves, representing potential robot trajectories. The effectiveness of our mapping and collision checking algorithms is evaluated in tasks requiring autonomous robot navigation and active mapping in unknown environments.
Data-driven approaches to Structural Health Monitoring (SHM) generally suffer from a lack of available health-state data. In particular, for most structures, it is not possible to obtain a comprehensive set of labelle...
详细信息
Data-driven approaches to Structural Health Monitoring (SHM) generally suffer from a lack of available health-state data. In particular, for most structures, it is not possible to obtain a comprehensive set of labelled damage data - even covering the most common damage types - due to impracticalities and economic considerations in observing the structure in a range of damage states. One solution to this problem is to utilise labelled data from a set of 'similar' structures. The assumption is that, as a population, the group may have a shared label set that covers a wider range of damage states, which can be used in labelling a different structure of interest. These goals, producing a model that generalises for a population of structures, and transferring label information between structures, are part of a population-based view of SHM - known as population-based SHM (PBSHM). By considering data from a population, it is possible to make data-driven SHM practical in industrial contexts beyond unsupervised learning, i.e. novelty detection. In order to realise the potential of PBSHM, this paper applies a heterogeneous transfer learning method - kernelised bayesian transfer learning (KBTL) - which is a sparsebayesian method that infers a discriminative classifier from inconsistent and heterogeneous feature data, i.e. the dataset from each member of the population may refer to different quantities in different dimensions. The technique infers a shared latent space where data from each member of the population are mapped on top of each other, meaning a single classifier can jointly be inferred that generalises to the complete population. As a consequence, label information can be transferred in this shared latent space between members of the population. The ability to infer a mapping from inconsistent and heterogeneous feature data make the approach a heterogeneous transfer learning method. To the best of the authors knowledge, this is the first time a heterogeneous transf
The background subtraction is a common method for real-time segmentation of moving targets in image sequences. This method needs a true image without moving objects. However, a background free of moving objects is not...
详细信息
The background subtraction is a common method for real-time segmentation of moving targets in image sequences. This method needs a true image without moving objects. However, a background free of moving objects is not available especially in traffic scene. The sparsebayesian algorithm combining optical flow and frame difference information for background initialization was proposed and proved its ability in improvement of the speed of initialing background. The inter-frame difference and optical flow value information is extracted as the input feature vector from multiple frame images including the moving objects. The sparsebayesian classifier is trained off-line. Then the background and moving object are partitioned by the classification probability exported from the trained classifier on-line. Thereby, the original background image can be constructed quickly. Experimental results are efficient for the background initialization.
暂无评论