Probabilistic Component Research Articles

Studies of reliability of power supply of the enterprise from the electric power system and autonomous loaded generators based on Markov processes

The purpose of this work is to substantiate the use of probabilistic components of the availa-bility factor of the power supply system (PSS) of the enterprise consisting of a electric power system and autonomous loaded generators, to calculate overpayments for electricity. To achieve this purpose, operation of the PSS during the last 5 years was studied. The PSS was considered as a restored system which has spinning reserve. The process of the PSS functioning was represented by a simple non-branching graph with one repair team and described by systems of differential and algebraic equations. The solution of the latter allowed calculating the probability components of the PSS availability factor. The operating time of the PSS in a calendar year was calculated with a different number of serviceable generators. The method for calculating overpayment for electricity was proposed. The most significant result is the justification of the use of probability components of the PSS availability factor for assessing overpayment for electricity. The significance of the obtained results lies in the fact that the proposed formula for overpayment for electricity allows to estimate overpayment, identify main causes and recommend measures to reduce it. The reliability of the results is confirmed by the example of calculating overpayment for electricity for a real enterprise. The newly obtained indicators of generator reliability can be used in the analysis of overpayment by other enterprises, which have similar generators.

A comparative study of different deep learning methods for time-series probabilistic residential load power forecasting

The widespread adoption of nonlinear power electronic devices in residential settings has significantly increased the stochasticity and uncertainty of power systems. The original load power data, characterized by numerous irregular, random, and probabilistic components, adversely impacts the predictive performance of deep learning techniques, particularly neural networks. To address this challenge, this paper proposes a time-series probabilistic load power prediction technique based on the mature neural network point prediction technique, i.e., decomposing the load power data into deterministic and stochastic components. The deterministic component is predicted using deep learning neural network technology, the stochastic component is fitted with Gaussian mixture distribution model and the parameters are fitted using great expectation algorithm, after which the stochastic component prediction data is obtained using the stochastic component generation method. Using a mature neural network point prediction technique, the study evaluates six different deep learning methods to forecast residential load power. By comparing the prediction errors of these methods, the optimal model is identified, leading to a substantial improvement in prediction accuracy.

Heterogeneous ensemble classifier in computer systems of medical diagnostics

The work is devoted to the solution of an actual scientific and technical problem tasks development of a support system for making diagnostic decisions in medicine based on the developed model of a heterogeneous ensemble classifier, which as basic models implements two approaches to development of diagnostic conclusions: probabilistic , based on the analysis of the training sample and expert information on the structure of symptom complexes of each disease. As a probabilistic component, the choice of the method of comparison with the standard, in which the diagnosed conditions of patients are represented by their standards in the space of signs, is justified. The geometric center of the class grouping in the space of diagnostic features is chosen as the benchmark for each class. Expert knowledge about the structure of the symptom complex is formalized by expressing the symptom complex of the disease in the form of numerical intervals of linguistic changes "below the norm", "norm", "above the norm". Various options for taking into account the opinion of experts about the structure of symptom complexes in the ensemble classifier are considered. Test verification of the developed classifier was conducted on real medical data, which confirmed its effectiveness.

From Proteomics to the Analysis of Single Protein Molecules.

Limit of detection (LoD) is a term that is used to characterize the sensitivity of an analytical method. The existing limitation of the sensitivity of analysis using modern mass spectrometry methods has been experimentally shown to be a limiting factor in the application of proteomic technologies in medicine. This article proposes a concept of a new technology that will set a new vector of development in the development of systems for solving problems of medical diagnostics and deals with theoretical and practical aspects of creating a new technology for the detection of single biomacromolecules (in particular, proteins) in biological samples. Such technology should be based on the principle of signal registration similar to that used in a Geiger counter (also known as a Geiger-Müller counter or G-M counter), a device that automatically counts the number of ionizing particles that hit it. This counter is free from probabilistic components; it registers a signal if there is at least one target molecule in the analysis chamber. Predictive medical diagnostics require technology based on methods where sensitivity allows for the detection of single marker molecules in a biological sample volume of 1-10 µL, the smallest volume of biomaterial used in laboratory diagnostics. Creation of a detector with a sensitivity of 10-18 M would allow for the detection of one molecule in 1 µL of the sample, which fundamentally makes this approach analogous to a G-M counter for solutions. To date, bioanalytical methods are limited to a sensitivity of 10-12 M (which is approximately 1 million molecules per 1 μL), which is insufficient to capture the early stages of pathological processes.

Merging two cultures: Deep and statistical learning

AbstractOur goal is to provide a review of deep learning methods which provide insight into structured high‐dimensional data. Merging the two cultures of algorithmic and statistical learning sheds light on model construction and improved prediction and inference, leveraging the duality and trade‐off between the two. Prediction, interpolation, and uncertainty quantification can be achieved using probabilistic methods at the output layer of the model. Rather than using shallow additive architectures common to most statistical models, deep learning uses layers of semi‐affine input transformations to provide a predictive rule. Applying these layers of transformations leads to a set of attributes (or, features) to which probabilistic statistical methods can be applied. Thus, the best of both worlds can be achieved: scalable prediction rules fortified with uncertainty quantification where sparse regularization finds the features. We review the duality between shallow and wide models such as principal components regression, and partial least squares and deep but skinny architectures such as autoencoders, multilayer perceptrons, convolutional neural net, and recurrent neural net. The connection with data transformations is of practical importance for finding good network architectures. By incorporating probabilistic components at the output level, the predictive uncertainty is allowed. We illustrate this idea by comparing plain Gaussian processes (GP) with partial least squares + Gaussian process (PLS + GP) and deep learning + Gaussian process (DL + GP).This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Deep Learning

A minute-resolution downscaling algorithm for high-resolution global irradiance time series

A minute-resolution downscaling algorithm for high-resolution global irradiance time series

POSSIBLE EVALUATION OF THE CORRECTNESS OF EXPLANATIONS TO THE END USER IN AN ARTIFICIAL INTELLIGENCE SYSTEM

The subject of this paper is the process of evaluation of explanations in an artificial intelligence system. The aim is to develop a method for forming a possible evaluation of the correctness of explanations for the end user in an artificial intelligence system. The evaluation of the correctness of explanations makes it possible to increase the user's confidence in the solution of an artificial intelligence system and, as a result, to create conditions for the effective use of this solution. Aims: to structure explanations according to the user's needs; to develop an indicator of the correctness of explanations using the theory of possibilities; to develop a method for evaluating the correctness of explanations using the possibilities approach. The approaches used are a set-theoretic approach to describe the elements of explanations in an artificial intelligence system; a possibility approach to provide a representation of the criterion for evaluating explanations in an intelligent system; a probabilistic approach to describe the probabilistic component of the evaluation of explanations. The following results are obtained. The explanations are structured according to the needs of the user. It is shown that the explanation of the decision process is used by specialists in the development of intelligent systems. Such an explanation represents a complete or partial sequence of steps to derive a decision in an artificial intelligence system. End users mostly use explanations of the result presented by an intelligent system. Such explanations usually define the relationship between the values of input variables and the resulting prediction. The article discusses the requirements for evaluating explanations, considering the needs of internal and external users of an artificial intelligence system. It is shown that it is advisable to use explanation fidelity evaluation for specialists in the development of such systems, and explanation correctness evaluation for external users. An explanation correctness assessment is proposed that uses the necessity indicator in the theory of possibilities. A method for evaluation of explanation fidelity is developed. Conclusions. The scientific novelty of the obtained results is as follows. A possible method for assessing the correctness of an explanation in an artificial intelligence system using the indicators of possibility and necessity is proposed. The method calculates the necessity of using the target value of the input variable in the explanation, taking into account the possibility of choosing alternative values of the variables, which makes it possible to ensure that the target value of the input variable is necessary for the explanation and that the explanation is correct.

Generative and Discriminative Learning for Lung X-Ray Analysis Based on Probabilistic Component Analysis.

The paper presents a hybrid generative/discriminative classification method aimed at identifying abnormalities, such as cancer, in lung X-ray images. The proposed method involves a generative model that performs generative embedding in Probabilistic Component Analysis (PrCA). The primary goal of PrCA is to model co-existing information within a probabilistic framework, with the intent to locate the feature vector space for X-ray data based on a defined kernel structure. A kernel-based classifier, grounded in information-theoretic principles, was employed in this study. The performance of the proposed method is evaluated against nearest neighbour (NN) classifiers and support vector machine (SVM) classifiers, which use a diagonal covariance matrix and incorporate normal linear and non-linear kernels, respectively. The method is found to achieve superior accuracy, offering a viable solution to the class of problems presented. Accuracy rates achieved by the kernels in the NN and SVM models were 95.02% and 92.45%, respectively, suggesting the method's competitiveness with state-of-the-art approaches.

Non-gendered pronoun processing: an investigation of the gender non-specific third person singular pronoun ‘TA’ in Chinese

ABSTRACT This study investigated the processing of the Chinese nongendered third-person singular pronoun, “TA,” in a series of self-paced reading experiments. We begin by investigating the perceived appropriateness of TA using a novel implementation of the modified maze task. We then contrasted reading latencies for TA and male- and female-gender pronouns in reference to antecedents with varying stereotypical gender (e.g., occupation terms) and definitional gender (e.g., kinship terms). In our analysis, we assessed several means of operationalizing stereotypical gender information. Optimal model performance was achieved with a continuous measure that accounted for individual differences in gender perception, suggesting the involvement of a probabilistic component. Results for reading latencies and perceived appropriateness of TA support previous findings from discourse analysis that TA is not entirely gender-neutral but rather has nuanced contexts of use in modern Chinese written discourse.

Why It Is Not Unreasonable to Fear Terrorism

ABSTRACTA common view has it that since we are far likelier to be killed in some road or household accident than in a terror attack, our fear of the latter is exaggerated. I argue that terrorism's relatively limited death toll need not mean that fearing it is unreasonable, nor does it immediately imply that counter‐terrorism policies are unjustified – whatever other, legitimate concerns these policies give rise to. First, I argue that in the special case of terrorism, it is misleading to focus on risk per capita, as critics typically do. Second, while terrorism has a probabilistic component which should be relevant to decision‐making, risk is not entirely or even primarily what terrorism is all about. Third, I argue that fearing terrorism may be reasonable even while recognizing the small probability of personal harm. Due to terrorism's random character, the belief that one will escape harm rests on little more than statistical evidence. As I explain, this leaves some room for reasonable doubt, and a justified level of fear.

Probabilistic reasoning for diagnosis prediction of Coronavirus disease based on probabilistic ontology

The novel Coronavirus has been declared a pandemic by the World Health Organization (WHO). Predicting the diagnosis of COVID-19 is essential for disease cure and control. The paper?s main aim is to predict the COVID-19 diagnosis using probabilistic ontologies to address the randomness and incompleteness of knowledge. Our approach begins with constructing the entities, attributes, and relationships of COVID-19 ontology, by extracting symptoms and risk factors. The probabilistic components of COVID-19 ontology are developed by creating a Multi-Entity Bayesian Network, then determining its components, with the different nodes, as probability distribution linked to various nodes. We use probabilistic inference for predicting COVID-19 diagnosis, using the Situation-Specific Bayesian Network (SSBN). To validate the solution, an experimental study is conducted on real cases, comparing the results of existing machine learning methods, our solution presents an encouraging result and, therefore enables fast medical assistance.

An Application of a Runtime Epistemic Probabilistic Event Calculus to Decision-making in e-Health Systems

AbstractWe present and discuss a runtime architecture that integrates sensorial data and classifiers with a logic-based decision-making system in the context of an e-Health system for the rehabilitation of children with neuromotor disorders. In this application, children perform a rehabilitation task in the form of games. The main aim of the system is to derive a set of parameters the child’s current level of cognitive and behavioral performance (e.g., engagement, attention, task accuracy) from the available sensors and classifiers (e.g., eye trackers, motion sensors, emotion recognition techniques) and take decisions accordingly. These decisions are typically aimed at improving the child’s performance by triggering appropriate re-engagement stimuli when their attention is low, by changing the game or making it more difficult when the child is losing interest in the task as it is too easy. Alongside state-of-the-art techniques for emotion recognition and head pose estimation, we use a runtime variant of a probabilistic and epistemic logic programming dialect of the Event Calculus, known as the Epistemic Probabilistic Event Calculus. In particular, the probabilistic component of this symbolic framework allows for a natural interface with the machine learning techniques. We overview the architecture and its components, and show some of its characteristics through a discussion of a running example and experiments.

TEMPORAL-ORIENTED MODEL OF CAUSAL RELATIONSHIP FOR CONSTRUCTING EXPLANATIONS FOR DECISION-MAKING PROCESS

The subject of research in the article is the decision-making process in intelligent systems. The goal is to develop a model of the causal relationship between the states of the decision-making process in an intelligent information system, taking into account the temporal aspect of this process, in order to build cause-and-effect relationships between the actions of the process and further use these dependencies to form explanations for the sequence of actions to obtain a decision. The formation of causal relations between the states of the decision-making process makes it possible to substantiate the sequence of actions of this process, considering incomplete information regarding external influences on this process. Tasks: structuring the decision-making process in an intelligent information system as a specialized business process; development of a three-element model of the causal relationship between the states of the decision-making process, considering the temporal aspect of this process; substantiation of the possibility of using three-element relationships to build causal dependencies for decision making in intelligent systems. The approaches used are: the set-theoretical approach used to describe the elements of the decision-making process in intelligent systems; a logical approach that provides a representation of the relationship between the states of the decision-making process; probabilistic approach to describe the probabilistic component of the decision-making process. The following results are obtained. The decision-making process in an intelligent information system was structured as a specialized business process that, using additional information from the user, turns the input data into a result that is valuable for this user; a three-element model of the causal relationship between the states of the decision-making process is proposed, which makes it possible to take into account external influences on the process; using a probabilistic approach, the possibility of using three-element causal relations to describe the decision-making process in intelligent systems is substantiated, taking into account uncontrolled external influences. Conclusions. The scientific novelty of the obtained results is as follows. A three-element model of the causal relationship between the states of the decision-making process is proposed, based on a model of a temporal rule of the "future" type, containing a state-cause, a state-effect and an intermediate state that reflects external influences. The model makes it possible to build a base of cause-and-effect dependencies for the decision-making process in an intelligent information system, considering external influences and use these dependencies to build explanations for this process.

The dynamics of ports hinterland formation: a probabilistic cellular automaton approach

This article presents a new methodology to determine the hinterland of a port as the aggregation of port choices made by interactive exporters. The main assumption is that exporters and ports interact in space and time between themselves, generating a complex system. Hinterland emerges as the aggregation of zones from which interacting agents send their shipments. This methodology builds upon the theory of cellular automata (CA), incorporating probabilistic components to allow for a rational representation of spatio-temporal exporters decisions. The CA's transition function includes neighbourhood interaction, network effects and ports attributes. The model's parameters are calibrated through a genetic algorithm. This methodology allows the estimation of ports demand capture as a time-dependent variable after a change in the ports level of service. This framework was applied to the automobile exports in Brazil, from 1997 to 2015, with a 73% of prediction success.

ENERGY EFFICIENCY OF COMBINED HEAT SUPPLY SYSTEMS

The consumption of energy resources in the world states is constantly growing from year to year. The production of fossil fuels is also increasing, but for various reasons it cannot fully cover the required amount from consumers. One of the most important consumption sectors is heat loads from heating, ventilation and hot water supply of industrial and residential buildings. To cover the thermal loads of heating and hot water supply, the necessary heat carrier is water heated to a certain temperature. The most promising from the point of view of heating water for hot water supply are solar collectors. Hot water for heating needs to be reheated practically throughout the entire heating period. The introduction of heat pumps is promising. When using solar collectors, the heating agent can be reheated in heat pumps.
 The aim of the study is to develop such a combined heat supply system that uses more renewable energy and as a peak source a fossil fuel boiler (electric energy), as well as a method for calculating this system to determine the optimal composition of equipment and rational modes of its operation.
 The methodology for calculating heat supply systems combining solar collectors, heat pumps and fossil fuel boilers is presented. The problem of load distribution between the main elements of the combined heat supply system should take into account the probabilistic component. This is due to the fact that with a changing real mode of operation, different thermal load of the equipment can be optimal. This is primarily influenced by the variable heat inputs from solar collectors during the day. According to the above method, it is possible to determine the optimal parameters of the heat supply system for different operating modes, at which the minimum consumption of fossil fuel will be ensured.

Neural implementations of Bayesian inference.

Bayesian inference has emerged as a general framework that captures how organisms make decisions under uncertainty. Recent experimental findings reveal disparate mechanisms for how the brain generates behaviors predicted by normative Bayesian theories. Here, we identify two broad classes of neural implementations for Bayesian inference: a modular class, where each probabilistic component of Bayesian computation is independently encoded and a transform class, where uncertain measurements are converted to Bayesian estimates through latent processes. Many recent experimental neuroscience findings studying probabilistic inference broadly fall into these classes. We identify potential avenues for synthesis across these two classes and the disparities that, at present, cannot be reconciled. We conclude that to distinguish among implementation hypotheses for Bayesian inference,we require greater engagement among theoretical and experimental neuroscientists in an effort that spans different scales of analysis, circuits, tasks, and species.

МОДЕЛИРОВАНИЕ СИСТЕМЫ «ПРОМЫСЕЛ» НА ОСНОВЕ БИОТЕХНОЛОГИЧЕСКОГО ДУАЛИЗМА

The article considers the essential feature of the system modeling of high-level fishery systems, e.g. the ‘Fishery’ system, the use of the principle of their internal self-organization, taking into account the influence of external factors - limitations. The internal structure of the ‘Fishery’ system is formed by a combination of biological, technical and technological components. Objects and sets of technical and technological components are interdependent, have a technical and technological nature and are combined into a single technological component. In this sense, the ‘Fishery’ system has a dual natural structure - biotechnological dualism, which is determined by the two components which simultaneously present and function in the system. In addition, the fishing process is greatly influenced by a set of restrictions and certain uncontrollable factors. By their simultaneous action they impart stochasticism and uncertainty to the fishing production process. Modeling of the ‘Fishery’ system was carried out using the decomposition method with consideration of the model of the composition of the system according to the signs (components) of duality - biological and technological components, which are its interrelated subsystems with the choice of an appropriate decomposition strategy - decomposition by subsystems (structural decomposition) with simultaneous embedding restrictive probabilistic component. Modeling of the blocks and the entire system was carried out taking into account the definition of component and restrictive relationships and relationships that determine both the static and dynamic nature of the ‘Fishery’ system in their initial state, i.e. without taking into account possible transient states in time intervals determined by the fishing situation. The generalized ‘Fishery’ system modeled on the basis of biotechnological dualism makes it possible to obtain a holistic view of its compositional composition, taking into account the impact of restrictions of a different nature, uncontrollable factors accompanying the production process of extraction of aquatic biological resources in the fishing areas. The resulting model in its initial static state is the initial biotechnological model for carrying out subsequent optimization control actions during its transition to a dynamic state in the time intervals indicated by the fishery

Interferometric Phase Reconstruction Based on Probability Generative Model: Toward Efficient Analysis of High-Dimensional SAR Stacks

In order to minimize the influence of decorrelation noise on multi-temporal interferometric synthetic aperture radar (MT-InSAR) applications, a series of phase reconstruction methods have been proposed in recent years. Unfortunately, current phase reconstruction methods generally exhibit a low computational efficiency due to their high non-linearity, in particular in the case that the dimension of a SAR stack is high. In this paper, a new approach is proposed to efficiently resolve phase reconstruction problems. This approach is inspired by the theory of probabilistic principle component analysis. A complex valued probability generative model is constructed to portray a phase reconstruction process. Moreover, in order to resolve such a model, a targeted algorithm based on the idea of expectation maximization is designed and implemented. For validation purposes, the proposed approach is compared to the traditional eigenvalue decomposition-based method by using simulated data and 101 real Sentinel-1A SAR images. The experimental results demonstrate that the proposed method can accelerate the phase reconstruction process drastically, in particular when a high-dimensional SAR stack is required to be processed.

Why it is Not Unreasonable to Fear Terrorism

The fear of terrorism is repeatedly scolded. A common claim is that since we are far more likely to be killed in some road or household accident than in a terror attack, the concern with the latter is unreasonably exaggerated. I argue that terrorism’s relatively limited death toll need not mean that fearing it is unreasonable, nor does it immediately imply that counter-terrorism policies are unjustified – whatever other, legitimate concerns these policies give rise to. First, in the special case of terrorism, it is misleading to focus on risk per capita, as critics typically do. Unlike some other hazards, the fact that an individual person is unlikely to be affected is not necessarily due to a diminished potential on the side of the terrorists. Rather, it may simply reflect the less important fact that the number of victims is expected to be small relative to the size of the population. Second, while terrorism has a probabilistic component which should be relevant to decision-making, risk is not entirely or even primarily what terrorism is all about. There are salient moral reasons for being concerned with terror – reasons that are irrelevant in the case of traffic- or household accidents. Third, I argue that fearing terrorism may be reasonable even while recognizing the small probability of personal harm. Due to terrorism’s random character, the belief that one will escape harm rests on little more than statistical evidence. As I explain, this evidence leaves room for reasonable doubt, which renders at least some level of fear reasonable as well. Please note: The reference list on the abstract page for this paper is automatically generated by a beta feature and may thus contain errors and omissions. Please consult the PDF version of the paper for complete and authoritative reference information.

A New Technique for Century-Scale Wind Component Indices

Advancing the understanding of how variations in the climate over the ocean influences the weather over the United States can be aided by developing marine climatic indices. Herein, wind component indices are developed using nearly 125 years of wind observations from ships. A new technique using probability density functions for the values of meridional and zonal wind components is developed to create indices for a user-selected region and accumulation interval (e.g., annual or seasonal) over a climatological period. The index is a measure of the shift in the likelihood of values above or below a threshold for a given season or year as compared to the long-term (e.g., 125 year) probability distribution. The new index method is demonstrated using ship-based wind observations for select regions of the Atlantic Ocean. Ship observations are extracted from release 3.0.0 of the International Comprehensive Ocean-Atmosphere Data Set. Prior to index creation, an assessment of wind data quality is completed, and suspect observations are removed. The method to create a probabilistic wind component index is described along with a metric of the uncertainty in the calculated index. Two wind component indices, for regions in the north Atlantic and eastern Gulf of Mexico, are presented to demonstrate the technique. Using the Gulf of Mexico index as a case study, we compare the wind component indices to precipitation measured over the Gulf coastal states and identify several relationships between multi-year changes in winds in the eastern Gulf of Mexico and precipitation on a seasonal basis. Exploring the spatiotemporal patterns of the onshore/offshore component wind indices derived from seasonal wind forecasts could provide a metric for future prediction of seasonal or annual precipitation to support the agricultural sector. The index method demonstrated can be applied to other spatiotemporal regions for different parameters and using other source datasets.