COVID-19 mortality rate prediction for India using statistical neural networks and gaussian process regression model

The primary purpose of this research is to identify the best COVID-19 mortality model for India using regression models and is to estimate the future COVID-19 mortality rate for India. Specifically, Statistical Neural Networks ( Radial Basis Function Neural Network (RBFNN), Generalized Regression Neural Network (GRNN)), and Gaussian Process Regression (GPR) are applied to develop the COVID-19 Mortality Rate Prediction (MRP) model for India. For that purpose, there are two types of dataset used in this study: One is COVID-19 Death cases, a Time Series Data and the other is COVID-19 Confirmed Case and Death Cases where Death case is dependent variable and the Confirmed case is an independent varia- ble. Hyperparameter optimization or tuning is used in these regression models, which is the process of identifying a set of optimal hyperparameters for any learning process with minimal error. Here, sigma (σ) is a hyperparameter whose value is used to constrain the learning process of the above models with minimum Root Mean Squared Error (RMSE). The perfor- mance of the models is evaluated using the RMSE and 'R2 values, which shows that the GRP model performs better than the GRNN and RBFNN. Keywords: Covid-19; India; mortality rate; mortality prediction; regression model; hyperparameter tuning; GPR; GRNN; RBFNN.

Gaussian synapses for probabilistic neural networks

The recent decline in energy, size and complexity scaling of traditional von Neumann architecture has resurrected considerable interest in brain-inspired computing. Artificial neural networks (ANNs) based on emerging devices, such as memristors, achieve brain-like computing but lack energy-efficiency. Furthermore, slow learning, incremental adaptation, and false convergence are unresolved challenges for ANNs. In this article we, therefore, introduce Gaussian synapses based on heterostructures of atomically thin two-dimensional (2D) layered materials, namely molybdenum disulfide and black phosphorus field effect transistors (FETs), as a class of analog and probabilistic computational primitives for hardware implementation of statistical neural networks. We also demonstrate complete tunability of amplitude, mean and standard deviation of the Gaussian synapse via threshold engineering in dual gated molybdenum disulfide and black phosphorus FETs. Finally, we show simulation results for classification of brainwaves using Gaussian synapse based probabilistic neural networks.

Detecting cloud contamination in passive microwave satellite measurements over land

Multiple geophysical parameters such as land surface temperature, are estimated using Microwave (MW) remote sensed brightness temperature. It is known that clouds do not affect those measurement in the MWs as much as in Visible and Infrared (VIS/IR), but some contamination can still occur when strong cloud formation (i.e. convective towers) or precipitation are present. To limit errors associated to cloud contamination in the estimation of surface parameters, we build an index giving the confidence to have an observation clear from contamination using standalone MW brightness temperature measurements. The method developed uses a statistical neural networks model built upon the Global Precipitation Microwave Imager (GPM-GMI) observations, with cloud presence information taken from Meteosat Third Generation-Spinning Enhanced Visible and Infrared Imager (MSG-SEVIRI). This index is available over land and ocean, and is developed for multiple frequency ranges to be applicable to successive generations of MW imagers (10 to 40 GHz, 10 to 100 GHz, 10 to 200 GHz). The index confidence increases with the number of channels available, and performs better over the ocean as expected. In all cases, even with a reduced number of information over land, the model reaches an accuracy > 70 %, in detecting contaminated observations. Finally an example application of this index to eliminate grid cells unsuitable for land surface temperature estimation is shown.