Brightness change is optimal stimulus for parasol retinal ganglion cells

Magnocellular-projecting retinal ganglion cells show spike response in two cases. Firstly, as a result of presentation of the optimal stimulus. Secondly, rebound excitation when removing the opposite stimulus. Also, there are studies suggesting that rebound excitation meets conditions to participate in visual perception at the same sensitivity and reaction speed as a response to the optimal stimulus. Thus, white noise stimulation creates possibility to catch the form of a smooth transition from one type of response to another. Using freely available data, a spike-triggered behavior map was built that does not show the area of silence between those two types of spike triggers. Moreover, linear filter with biphasic temporal properties which work as the derivative kernel demonstrate that both responses are two sides of the same coin. Thus, it is suggested to determine the optimal stimulus for magnocellular-projecting retinal ganglion cells as brightness change according to concentric center–surround receptive field structure.

The Distributed and Centralized Fusion Filtering Problems of Tessarine Signals from Multi-Sensor Randomly Delayed and Missing Observations under Tk-Properness Conditions

This paper addresses the fusion estimation problem in tessarine systems with multi-sensor observations affected by mixed uncertainties when under Tk-properness conditions. Observations from each sensor can be updated, delayed, or contain only noise, and a correlation is assumed between the state and the observation noises. Recursive algorithms for the optimal local linear filter at each sensor as well as both centralized and distributed linear fusion estimators are derived using an innovation approach. The Tk-properness assumption implies a reduction in the dimension of the augmented system, which yields computational savings in the previously mentioned algorithms compared to their counterparts, which are derived from real or widely linear processing. A numerical simulation example illustrates the obtained theoretical results and allows us to visualize, among other aspects, the insignificant difference in the accuracy of both fusion filters, which means that the distributed filter, although suboptimal, is preferable in practice as it implies a lower computational cost.

Gaussian Processes for Regression and Optimisation

<p>Gaussian processes have proved to be useful and powerful constructs for the purposes of regression. The classical method proceeds by parameterising a covariance function, and then infers the parameters given the training data. In this thesis, the classical approach is augmented by interpreting Gaussian processes as the outputs of linear filters excited by white noise. This enables a straightforward definition of dependent Gaussian processes as the outputs of a multiple output linear filter excited by multiple noise sources. We show how dependent Gaussian processes defined in this way can also be used for the purposes of system identification. Onewell known problem with Gaussian process regression is that the computational complexity scales poorly with the amount of training data. We review one approximate solution that alleviates this problem, namely reduced rank Gaussian processes. We then show how the reduced rank approximation can be applied to allow for the efficient computation of dependent Gaussian processes. We then examine the application of Gaussian processes to the solution of other machine learning problems. To do so, we review methods for the parameterisation of full covariance matrices. Furthermore, we discuss how improvements can be made by marginalising over alternative models, and introduce methods to perform these computations efficiently. In particular, we introduce sequential annealed importance sampling as a method for calculating model evidence in an on-line fashion as new data arrives. Gaussian process regression can also be applied to optimisation. An algorithm is described that uses model comparison between multiple models to find the optimum of a function while taking as few samples as possible. This algorithm shows impressive performance on the standard control problem of double pole balancing. Finally, we describe how Gaussian processes can be used to efficiently estimate gradients of noisy functions, and numerically estimate integrals.</p>

Gaussian Processes for Regression and Optimisation

<p>Gaussian processes have proved to be useful and powerful constructs for the purposes of regression. The classical method proceeds by parameterising a covariance function, and then infers the parameters given the training data. In this thesis, the classical approach is augmented by interpreting Gaussian processes as the outputs of linear filters excited by white noise. This enables a straightforward definition of dependent Gaussian processes as the outputs of a multiple output linear filter excited by multiple noise sources. We show how dependent Gaussian processes defined in this way can also be used for the purposes of system identification. Onewell known problem with Gaussian process regression is that the computational complexity scales poorly with the amount of training data. We review one approximate solution that alleviates this problem, namely reduced rank Gaussian processes. We then show how the reduced rank approximation can be applied to allow for the efficient computation of dependent Gaussian processes. We then examine the application of Gaussian processes to the solution of other machine learning problems. To do so, we review methods for the parameterisation of full covariance matrices. Furthermore, we discuss how improvements can be made by marginalising over alternative models, and introduce methods to perform these computations efficiently. In particular, we introduce sequential annealed importance sampling as a method for calculating model evidence in an on-line fashion as new data arrives. Gaussian process regression can also be applied to optimisation. An algorithm is described that uses model comparison between multiple models to find the optimum of a function while taking as few samples as possible. This algorithm shows impressive performance on the standard control problem of double pole balancing. Finally, we describe how Gaussian processes can be used to efficiently estimate gradients of noisy functions, and numerically estimate integrals.</p>

A Hydrological Digital Twin by Artificial Neural Networks for Flood Simulation in Gardon de Sainte-Croix Basin, France

Understanding, simulating and forecasting dynamic and nonlinear natural phenomena are necessary in a climate change context and increased sensitivity of societies to natural hazards. Nevertheless, even though powerful computing tools and algorithms have been widely used to understand and to predict natural disasters, these tasks are still challenging for scientists. Indeed, one of the most dangerous natural phenomena, flash floods keep being a challenge for modelers, despite (i) the existence of some effective hydrological simulating tools, and (ii) the increasing availability of descriptive data, especially rainfall and discharge. In particular, on one hand, environmental data contain an important amount of noise leading to additional uncertainties and on the other hand, physically based models strongly depend on assumptions about the behavior of the basin, that is often more variable in space and time than what is modelled. With the objective of applying data assimilation to improve forecasting properties of the physical model, it is necessary to dispose of a differentiable model. In order to mitigate this issue, a hybrid physical and statistical approach is proposed in this study. It was shown in previous works that deep neural networks are able to identify any differentiable function by using the universal approximation property. Deep neural networks are also good candidates to perform the digital twin of the physical model. Thus, three different neural networks models were designed in this study, and each one is implementing a different type of non-linear filter model, in order to achieve the dynamic character of the catchment area (recurrent, feedforward and static models). The study area is located in the Gardon de Sainte-Croix basin (France), which is known for its sudden and violent floods that caused casualties and a lot of damage. The chosen physical-based model is semi distributed conceptual hydrological SOCONT model, RS Minerve (https://www.crealp.ch/down/rsm/install2/archives.html). Neural networks design was done by using a rigorous complexity selection and regularization methods to promote a good generalization. The three models obtained were thus compared. The feed forward model gave the best results on tests events (Nash score=0.98−0.99), making full use of the inputs with previous observed discharges whereas the recurrent model gave interesting results representing satisfactorily the dynamics of the physical model (Nash score=0.8−0.97). The static model, whose inputs contain only rainfall, is less efficient, showing the importance of dynamics in that kind of system (Nash score=0.62−0.84). Beyond data assimilation, these results open paths of inquiry for building digital twins of physical model, allowing also a great reduction of computing time.

Identification of maneuvering objects during structural and system control of airspace

Работа посвящена построению процедуры идентификации маневрирующих объектов с использованием критерия идеального наблюдателя и фильтрации параметров трасс при сопровождении средствами мониторинга в интересах структурносистемного контроля воздушного пространства. Для минимизации среднеквадратических ошибок оценок координат и скоростей движения объектов разработаны алгоритмы экстраполяции параметров траекторий путем задания корректирующего шумового ускорения и замены результатов фильтрации оценок координат на измеренные значения при распознавании маневра. Обоснованы параметры фильтрации с шумовым ускорением в зависимости от точности измерений пространственных характеристик и идентификации при группировании однотипных признаков с наибольшими значениями условных вероятностей ситуаций отождествления объектов Purpose. This work addresses construction of the procedure for identifying maneuvering air objects in the process of tracking their routes. Monitoring tools during structural and system air space control are employed. The study is aimed to establish the abilities of correct identification of objects and false alarm at various standard errors of measurements of angular coordinates and to determine ways to increase efficiency of identifications performed due to selection of filtering options during trace tracking. Methodology. Identification of objects was performed according to the ideal observer criterion by comparing estimates of angular coordinates of objects subjected to linear filtering with corrective noise acceleration. In order to minimize root-mean-square errors of coordinates and motion velocity estimates of objects, route parameter extrapolation algorithms are obtained by setting correcting noise acceleration and replacing the results of filtering coordinate estimates with measured values during manoeuvre recognition. Due to a priori uncertainty of route parameters, target tracking was initially performed using recurring linear filtering while maintaining the priority of straight uniform movement. The recognition of the maneuver was carried out as a result of exceeding the difference between the measured and filtered values of the target coordinates of the threshold value. Findings. Filtering parameters with noise acceleration are justified depending on the accuracy of measurements of spatial characteristics and identification when grouping identification features with the highest values of conditional probabilities of situations for the objects under identification. As a result of replacing filtered parameters of alignments containing areas with rotations of 10 and 20, measured values for standard bearing errors (1 ... 2), the maximum error in determining directions for objects reaches 0.8 and 0.9, respectively. When replacing the estimates of the parameters of the alignments obtained using a recurring linear filter without taking into account noise acceleration, the coordinate values measured at the bearing error (0 . 5 ... 2), the errors of the filtered bearing of the targets at the angles of rotation of 10are (0 . 2 ... 1). When maneuvering objects with turns by 20, the largest value of the standard bearing error increases to 1.2. By increasing the accuracy of the diaper from 2 to 0.5, the probability of correct identification of objects in monitoring tools performing noise correction acceleration filtering increases by about 3 times and reaches a value of 0.9. As a result of replacing the estimates of the parameters of the alignments filtered taking into account the corrective noise acceleration with the results of measurements, the probability of correct identification of objects with standard bearing errors of not more than 0.5decreases from 0.9 to 0.85. Originality/value. The identification of maneuvering air objects is performed using filtering of route parameters calculated with the help of the ideal observer criterion. For the most efficient identification, the identification features belonging to the same object must be established according to the highest conditional probability of the identification situation. To minimize errors in estimation of the angular coordinates of objects, a procedure for filtering motion parameters with corrective noise acceleration is implemented

An Adaptive Linear Filter model of procedural category learning

We use a feature-based association model to fit grouped and individual level category learning and transfer data. The model assumes that people use corrective feedback to learn individual feature to categorization-criterion correlations and combine those correlations additively to produce classifications. The current model is an Adaptive Linear Filter (ALF) with logistic output function and Least Mean Squares learning algorithm. Categorization probabilities are computed by a logistic function. Our data spans over 31 published data sets. Both at grouped and individual level analysis levels, the model performs remarkably well, accounting for large amounts of available variances. Notably, the model achieves its fits with a very minimal number of free parameters. We discuss the ALF’s advantages as a model of procedural categorization, in terms of its simplicity, its ability to capture empirical trends and its ability to solve longstanding challenges to similar models. In particular, we discuss why the model is not equivalent to a prototype model, as previously thought.

Modeling and Analysis of Data Prediction Technique Based on Linear Regression Model (DP-LRM) for Cluster-Based Sensor Networks

Recent developments in information gathering procedures and the collection of big data over a period of time as a result of introducing high computing devices pose new challenges in sensor networks. Data prediction has emerged as a key area of research to reduce transmission cost acting as principle analytic tool. The transformation of huge amount of data into an equivalent reduced dataset and maintaining data accuracy and integrity is the prerequisite of any sensor network application. To overcome these challenges, a data prediction technique is suggested to reduce transmission of redundant data by developing a regression model on linear descriptors on continuous sensed data values. The proposed model addresses the basic issues involved in data aggregation. It uses a buffer based linear filter algorithm which compares all incoming values and establishes a correlation between them. The cluster head is accountable for predicting data values in the same time slot, calculates the deviation of data values, and propagates the predicted values to the sink.

Efficient Filtering for Edge Extraction under Perspective Effect

Though it is generally believed that edges should be extracted at different scales when using a linear filter, it is still difficult to determine the optimal scale for each filter. In this paper, we propose a novel approach called orientation and scale tuned difference of boxes (osDoB) to solve this problem. For certain computer vision applications, such as lane marking detection, the prior information about the concerned target can facilitate edge extraction in a top-down manner. Based on the perspective effect, we associate the scale of the edge in an image with the target size in the real world and assign orientation and scale parameters for filtering each pixel. Considering the fact that it is very time-consuming to naïvely perform filters with different orientations and scales, we further design an extended integration map technology to speed up filtering. Our method is validated on synthetic and real data. The experimental results show that assigning appropriate orientation and scale parameters for filters is effective and can be realized efficiently.