Explore the vast range of titles available.

Over the past decade fMRI researchers have developed increasingly sensitive techniques for analyzing the information represented in BOLD activity. The most popular of these techniques is linear classification, a simple technique for decoding information about experimental stimuli or tasks from patterns of activity across an array of voxels. A more recent development is the voxel-based encoding model, which describes the information about the stimulus or task that is represented in the activity of single voxels. Encoding and decoding are complementary operations: encoding uses stimuli to predict activity while decoding uses activity to predict information about the stimuli. However, in practice these two operations are often confused, and their respective strengths and weaknesses have not been made clear. Here we use the concept of a linearizing feature space to clarify the relationship between encoding and decoding. We show that encoding and decoding operations can both be used to investigate some of the most common questions about how information is represented in the brain. However, focusing on encoding models offers two important advantages over decoding. First, an encoding model can in principle provide a complete functional description of a region of interest, while a decoding model can provide only a partial description. Second, while it is straightforward to derive an optimal decoding model from an encoding model it is much more difficult to derive an encoding model from a decoding model. We propose a systematic modeling approach that begins by estimating an encoding model for every voxel in a scan and ends by using the estimated encoding models to perform decoding. ►Encoding and decoding can be described in terms of a linearizing feature space. ►Encoding models can provide complete functional descriptions; decoding models cannot. ►Decoding models can relate activity directly to behavior; encoding models cannot. ►An encoding model is more easily converted to a decoding model than vice versa.

Accurate detection of road surface conditions in adverse winter weather is essential for traffic safety. To promote safe driving and efficient road management, this study presents an accurate and generalizable data-driven learning model for the estimation of road surface conditions. The machine model was a support vector machine (SVM), which has been successfully applied in diverse fields, and kernel functions (linear, Gaussian, second-order polynomial) with a soft margin classification technique were also adopted. Two learner designs (one-vs-one, one-vs-all) extended their application to multi-class classification. In addition to this non-probabilistic classifier, this study calculated the posterior probability of belonging to each group by applying the sigmoid function to the classification scores obtained by the trained SVM. The results indicate that the classification errors of all the classifiers, excluding the one-vs-all linear learners, were below 3%, thereby accurately classifying road surface conditions, and that the generalization performance of all the one-vs-one learners was within an error rate of 4%. The results also showed that the posterior probabilities can analyze certain atmospheric and road surface conditions that correspond to a high probability of hazardous road surface conditions. Therefore, this study demonstrates the potential of data-driven learning models in classifying road surface conditions accurately.

Land Use and Land Cover (LULC) classification is critical for monitoring and managing natural resources and urban development. This study focuses on LULC classification for change detection analysis of remotely sensed data using a machine learning-based Random Forest classifier. The research aims to provide a detailed analysis of LULC changes between 2010 and 2020. The Random Forest classifier is chosen for its robustness and high accuracy in handling complex datasets. The classifier achieved a classification accuracy of 86.56% for the 2010 data and 88.42% for the 2020 data, demonstrating an improvement in classification performance over the decade. The results indicate significant LULC changes, highlighting areas of urban expansion, deforestation, and agricultural transformation. These findings highlight the importance of continuous monitoring and provide valuable insights for policymakers and environmental managers. The study demonstrates the effectiveness of using advanced machine-learning techniques for accurate LULC classification and change detection in remotely sensed data.

Even within the early sensory areas, the majority of the input to any given cortical neuron comes from other cortical neurons. To extend our knowledge of the contextual information that is transmitted by such lateral and feedback connections, we investigated how visually nonstimulated regions in primary visual cortex (V1) and visual area V2 are influenced by the surrounding context. We used functional magnetic resonance imaging (fMRI) and pattern-classification methods to show that the cortical representation of a nonstimulated quarter-field carries information that can discriminate the surrounding visual context. We show further that the activity patterns in these regions are significantly related to those observed with feed-forward stimulation and that these effects are driven primarily by V1. These results thus demonstrate that visual context strongly influences early visual areas even in the absence of differential feed-forward thalamic stimulation.

Least squares estimate that can directly obtain the analytical solution to minimize the mean square error (MSE) is one of the most effective regression tools. It has also been applied to many classification tasks due to its simplicity, clear physical and tractability. However, there is a fundamental contradiction between regression with continuous samples and classification with discrete category labels, i.e., the output metric spaces are different. Considering this contradiction in regression-based classification, this paper presents a new linear classifier, termed penalized least squares classifier (PLSC), which gradually adds the penalty on the distance between misclassified samples and decision boundary to the loss function. Fundamentally, the decision boundary is obtained by minimizing MSE with iterative cost-sensitive learning. Then, an enhanced nonlinear neural network classifier, PLSC-BP is formulated, in which the penalized least squares mechanism is applied to tune the learning strategy of the conventional neural network by adjusting the cost factor for each sample. Extensive experiments on six synthetic datasets and eleven publicly available datasets show that classification, which cannot be realized by traditional least squares, can be conducted using a regression algorithm with iterative cost-sensitive learning. The proposed enhanced algorithm outperforms the traditional neural network classifiers and other classifiers in terms of classification.

Although linear classifiers are one of the oldest methods in machine learning, they are still very popular in the machine learning community. This is due to their low computational complexity and robustness to overfitting. Consequently, linear classifiers are often used as base classifiers of multiple ensemble classification systems. This research is aimed at building a new fusion method dedicated to the ensemble of linear classifiers. The fusion scheme uses both measurement space and geometrical space. Namely, we proposed a probability-driven scoring function which shape depends on the orientation of the decision hyperplanes generated by the base classifiers. The proposed fusion method is compared with the reference method using multiple benchmark datasets taken from the KEEL repository. The comparison is done using multiple quality criteria. The statistical analysis of the obtained results is also performed. The experimental study shows that, under certain conditions, some improvement may be obtained.

For logistic regression with massive data, subsampling is an effective way to alleviate the computational challenge. In contrast to most existing methods in the literature that select subsamples randomly, we propose to obtain subsamples in a deterministic way. To be more specific, we measure with leverage scores the influence of each sample to model fitting and select the ones with the highest scores deterministically. We propose a faster alternative method by mimicking the leverage scores with a simple and intuitive form. Our methods pick subsamples catering for constructing a linear classification boundary and hence are more efficient when the subsample size is small. We derive non-asymptotic properties of the two methods regarding the observed information, prediction, and parameter estimation accuracy. Extensive simulation studies and two real applications validate the theoretical results and demonstrate the superiority of our methods.

Dimension reduction is an important, often essential, component of multivariate analysis. It is usually an intermediary step to make the multivariate inference of interest meaningful. In classification, for example, it helps determine the subset of features that significantly contribute to classification, in order to enhance the optimality of the classifier. The present article addresses this issue by modifying a classical test to determine the feature subspace which may be discarded as redundant to enable the remaining, potentially contributing, features to improve the classifier. The proposed test allows the dimension, also of sub-vectors, to exceed the sample sizes. The test is constructed under a general multivariate model, with normality as a special case, and a few mild assumptions. Two-class case is discussed in detail, with a brief extension to multi-class case. Simulations are used to demonstrate the accuracy of the proposed theory, and its applications are illustrated through several real data examples.

Linear and mixed binary programming techniques, which arise from model linearity, are widely supported by advanced optimization solvers. In this paper, we present a comprehensive guide for transforming linear classifiers into linear or mixed binary programming tasks. Our approach employs widely used techniques, such as Kesler construction with either perfect or imperfect learning, and weight regularization using the L1 or L0 norm, enhanced by additional maximum weight constraint (i.e., L∞). Various linear optimization tasks are formulated based on performance measures such as accuracy and sensitivity. The classifiers are constructed using different weight penalizations and regularizations – specifically, the L0 norm, which yields mixed binary programming tasks with NP-hard complexity, and the L1 norm, which results in linear programming tasks with polynomial complexity, both with an additional maximum weight constraint. The proposed classifiers are compared on several UCI datasets (Iris Flower, Wine, and Seeds) and match or outperform Ridge and Lasso regression methods when applied to classification tasks.

As a special form of piecewise linear classifier, the convex polyhedron classifier is simple to implement and achieves rapid response in real-time classification. However, it usually performs badly in the case of high noise where severe boundary intrusion exists. Inspired by the scheme of soft margin in support vector machine, in this paper we propose a soft-margin convex polyhedron classifier for nonlinear classification task. The base (linear) classifier is first generalized to its soft-margin version through kernelization process and slack variables. In each local region, the soft-margin base classifier learns a decision hyperplane with noise tolerance. Then, a series of learned hyperplanes are structurally integrated into a convex polyhedron classifier, which is essentially a convex polyhedron that encloses one class and excludes the other class outside. Experimental results on fifteen benchmark datasets show the proposed soft-margin convex polyhedron classifier is comparable to linear support vector machine and four piecewise linear classifiers, but does not perform as well as the support vector machine with radial basis function kernel in general. When random noises are added to datasets, the soft-margin convex polyhedron classifier achieves similar or better accuracies with the well-known classifiers used for comparison, implying its promising ability of noise tolerance.