Cortical ß Power Reflects a Neural Implementation of Decision Boundary Collapse in Speeded Decisions.

A prominent account of decision-making assumes that information is accumulated until a fixed response threshold is crossed. However, many decisions require weighting of information appropriately against time. Collapsing response thresholds are a mathematically optimal solution to this decision problem. However, our understanding of the neurocomputational mechanisms underlying dynamic response thresholds remains significantly incomplete. To investigate this issue, we used a multistage drift-diffusion model (DDM) and also analyzed EEG ß power lateralization (BPL). The latter served as a neural proxy for decision signals. We analyzed a large dataset (n = 863; 434 females and 429 males) from a speeded flanker task and data from an independent confirmation sample (n = 119; 70 females and 49 males). We showed that a DDM with collapsing decision thresholds, a process wherein the decision boundary reduces over time, captured participants' time-dependent decision policy more accurately than a model with fixed thresholds. Previous research suggests that BPL over motor cortices reflects features of a decision signal and that its peak, coinciding with the motor response, may serve as a neural proxy for the decision threshold. We show that BPL around the response decreased with increasing RTs. Together, our findings offer compelling evidence for the existence of collapsing decision thresholds in decision-making processes.

Developing probabilistic ensemble machine learning models for home-based sleep apnea screening using overnight SpO2 data at varying data granularity.

PURPOSE: This study aims to develop sleep apnea screening models with overnight SpO2 data, and to investigate the impact of the SpO2 data granularity on model performance. METHODS: A total of 7,718 SpO2 recordings from the SHHS and MESA datasets were used. Probabilistic ensemble machine learning was employed to predict sleep apnea status at three AHI cutoff points: ≥ 5, ≥ 15, and ≥ 30 events/hour. To investigate the impact of data granularity, SpO2 data were aggregated at 30, 60, and 300 s. RESULTS: Our models demonstrated good to excellent performance on internal test, with average area under the curve (AUC) values of 0.91, 0.93, and 0.96 for cutoffs ≥ 5, ≥ 15, and ≥ 30 at data granularity of 1 s, respectively. Both sensitivity (0.76, 0.84, 0.89) and specificity (0.87, 0.86, 0.90) ranged from good to excellent across three cutoffs. Positive predictive values (PPV) ranged from excellent to fair (0.97, 0.83, 0.66), and negative predictive values (NPV) ranged from low to excellent (0.43, 0.87, 0.98). Model performance on external test slightly dropped compared to internal test, but still achieved good to excellent AUC above 0.80 across all data granularity and all the three cutoffs. Data granularity of 300 s led to a reduction in performance metrics across all cutoffs. CONCLUSION: Our models demonstrated superior performance across all three AHI cutoff thresholds compared to existing large sleep apnea screening models, even when considering varying SpO2 data granularity. However, lower data granularity was associated with decreased screening performance, indicating a need for further research in this area.

MESMERIC: Machine Learning-Based Trust Management Mechanism for the Internet of Vehicles.

The emerging yet promising paradigm of the Internet of Vehicles (IoV) has recently gained considerable attention from researchers from academia and industry. As an indispensable constituent of the futuristic smart cities, the underlying essence of the IoV is to facilitate vehicles to exchange safety-critical information with the other vehicles in their neighborhood, vulnerable pedestrians, supporting infrastructure, and the backbone network via vehicle-to-everything communication in a bid to enhance the road safety by mitigating the unwarranted road accidents via ensuring safer navigation together with guaranteeing the intelligent traffic flows. This requires that the safety-critical messages exchanged within an IoV network and the vehicles that disseminate the same are highly reliable (i.e., trustworthy); otherwise, the entire IoV network could be jeopardized. A state-of-the-art trust-based mechanism is, therefore, highly imperative for identifying and removing malicious vehicles from an IoV network. Accordingly, in this paper, a machine learning-based trust management mechanism, MESMERIC, has been proposed that takes into account the notions of direct trust (encompassing the trust attributes of interaction success rate, similarity, familiarity, and reward and punishment), indirect trust (involving confidence of a particular trustor on the neighboring nodes of a trustee, and the direct trust between the said neighboring nodes and the trustee), and context (comprising vehicle types and operating scenarios) in order to not only ascertain the trust of vehicles in an IoV network but to segregate the trustworthy vehicles from the untrustworthy ones by means of an optimal decision boundary. A comprehensive evaluation of the envisaged trust management mechanism has been carried out which demonstrates that it outperforms other state-of-the-art trust management mechanisms.

Understanding the Feature Space and Decision Boundaries of Commercial WAFs Using Maximum Entropy in the Mean.

The security of a network requires the correct identification and characterization of the attacks through its ports. This involves the follow-up of all the requests for access to the networks by all kinds of users. We consider the frequency of connections and the type of connections to a network, and determine their joint probability. This leads to the problem of determining a joint probability distribution from the knowledge of its marginals in the presence of errors of measurement. Mathematically, this consists of an ill-posed linear problem with convex constraints, which we solved by the method of maximum entropy in the mean. This procedure is flexible enough to accommodate errors in the data in a natural way. Also, the procedure is model-free and, hence, it does not require fitting unknown parameters.

Single-trial dynamics explain magnitude sensitive decision making.

BACKGROUND: Previous research has reported or predicted, on the basis of theoretical and computational work, magnitude sensitive reaction times. Magnitude sensitivity can arise (1) as a function of single-trial dynamics and/or (2) as recent computational work has suggested, while single-trial dynamics may be magnitude insensitive, magnitude sensitivity could arise as a function of overall reward received which in turn affects the speed at which decision boundaries collapse, allowing faster responses as the overall reward received increases. RESULTS: Here, we review previous theoretical and empirical results and we present new evidence for magnitude sensitivity arising as a function of single-trial dynamics. CONCLUSIONS: The result of magnitude sensitive reaction times reported is not compatible with single-trial magnitude insensitive models, such as the statistically optimal drift diffusion model.

Distance-based novelty detection model for identifying individuals at risk of developing Alzheimer's disease.

Introduction: Novelty detection (ND, also known as one-class classification) is a machine learning technique used to identify patterns that are typical of the majority class and can discriminate deviations as novelties. In the context of Alzheimer's disease (AD), ND could be employed to detect abnormal or atypical behavior that may indicate early signs of cognitive decline or the presence of the disease. To date, few research studies have used ND to discriminate the risk of developing AD and mild cognitive impairment (MCI) from healthy controls (HC). Methods: In this work, two distinct cohorts with highly heterogeneous data, derived from the Australian Imaging Biomarkers and Lifestyle (AIBL) Flagship Study of Ageing project and the Fujian Medical University Union Hospital (FMUUH) China, were employed. An innovative framework with built-in easily interpretable ND models constructed solely on HC data was introduced along with proposing a strategy of distance to boundary (DtB) to detect MCI and AD. Subsequently, a web-based graphical user interface (GUI) that incorporates the proposed framework was developed for non-technical stakeholders. Results: Our experimental results indicate that the best overall performance of detecting AD individuals in AIBL and FMUUH datasets was obtained by using the Mixture of Gaussian-based ND algorithm applied to single modality, with an AUC of 0.8757 and 0.9443, a sensitivity of 96.79% and 89.09%, and a specificity of 89.63% and 90.92%, respectively. Discussion: The GUI offers an interactive platform to aid stakeholders in making diagnoses of MCI and AD, enabling streamlined decision-making processes. More importantly, the proposed DtB strategy could visually and quantitatively identify individuals at risk of developing AD.

Hard Boundary-Based Neurofeedback Training Procedure: A Modified Fixed Thresholding Method for More Accurate Guidance of Subjects Within Target Areas During Neurofeedback Training.

In nearly all studies within the domain of neurofeedback, a threshold has been defined for each training feature in a way that subjects' status can be evaluated during training according to the given value. In this study, a hard boundary-based neurofeedback training (HBNFT) method based on the determination of decision boundary using support vector machine (SVM) classifier was proposed in which subjects' status were clarified considering a decision boundary and they could also be encouraged once entering a target area. In this method, a scoring index (SI) was similarly defined whose value was determined in accordance with subject performance during training. The results revealed that employing a classifier and determining a decision boundary instead of using a threshold could prove more successful in accurately guiding them towards a target area and also meet no needs to choose a basis for determining a threshold. Moreover, it was likely that the proposed method could be more efficient in controlling features and preventing extreme changes compared to those using variable thresholds.

Probabilistic machine learning for breast cancer classification.

A probabilistic neural network has been implemented to predict the malignancy of breast cancer cells, based on a data set, the features of which are used for the formulation and training of a model for a binary classification problem. The focus is placed on considerations when building the model, in order to achieve not only accuracy but also a safe quantification of the expected uncertainty of the calculated network parameters and the medical prognosis. The source code is included to make the results reproducible, also in accordance with the latest trending in machine learning research, named Papers with Code. The various steps taken for the code development are introduced in detail but also the results are visually displayed and critically analyzed also in the sense of explainable artificial intelligence. In statistical-classification problems, the decision boundary is the region of the problem space in which the classification label of the classifier is ambiguous. Problem aspects and model parameters which influence the decision boundary are a special aspect of practical investigation considered in this work. Classification results issued by technically transparent machine learning software can inspire more confidence, as regards their trustworthiness which is very important, especially in the case of medical prognosis. Furthermore, transparency allows the user to adapt models and learning processes to the specific needs of a problem and has a boosting influence on the development of new methods in relevant machine learning fields (transfer learning).

Feature Augmentation via Nonparametrics and Selection (FANS) in High-Dimensional Classification.

We propose a high dimensional classification method that involves nonparametric feature augmentation. Knowing that marginal density ratios are the most powerful univariate classifiers, we use the ratio estimates to transform the original feature measurements. Subsequently, penalized logistic regression is invoked, taking as input the newly transformed or augmented features. This procedure trains models equipped with local complexity and global simplicity, thereby avoiding the curse of dimensionality while creating a flexible nonlinear decision boundary. The resulting method is called Feature Augmentation via Nonparametrics and Selection (FANS). We motivate FANS by generalizing the Naive Bayes model, writing the log ratio of joint densities as a linear combination of those of marginal densities. It is related to generalized additive models, but has better interpretability and computability. Risk bounds are developed for FANS. In numerical analysis, FANS is compared with competing methods, so as to provide a guideline on its best application domain. Real data analysis demonstrates that FANS performs very competitively on benchmark email spam and gene expression data sets. Moreover, FANS is implemented by an extremely fast algorithm through parallel computing.

Near-Bayesian Support Vector Machines for imbalanced data classification with equal or unequal misclassification costs.

Support Vector Machines (SVMs) form a family of popular classifier algorithms originally developed to solve two-class classification problems. However, SVMs are likely to perform poorly in situations with data imbalance between the classes, particularly when the target class is under-represented. This paper proposes a Near-Bayesian Support Vector Machine (NBSVM) for such imbalanced classification problems, by combining the philosophies of decision boundary shift and unequal regularization costs. Based on certain assumptions which hold true for most real-world datasets, we use the fractions of representation from each of the classes, to achieve the boundary shift as well as the asymmetric regularization costs. The proposed approach is extended to the multi-class scenario and also adapted for cases with unequal misclassification costs for the different classes. Extensive comparison with standard SVM and some state-of-the-art methods is furnished as a proof of the ability of the proposed approach to perform competitively on imbalanced datasets. A modified Sequential Minimal Optimization (SMO) algorithm is also presented to solve the NBSVM optimization problem in a computationally efficient manner.