Spatially-resolved hyperlocal weather prediction and anomaly detection using IoT sensor networks and machine learning techniques

Accurate and timely hyperlocal weather predictions are essential for various applications, ranging from agriculture to disaster management. In this paper, we propose a novel approach that combines hyperlocal weather prediction and anomaly detection using IoT sensor networks and advanced machine learning techniques. Our approach leverages data from multiple spatially-distributed yet relatively close locations and IoT sensors to create high-resolution weather models capable of predicting short-term, localized weather conditions such as temperature, pressure, and humidity. By monitoring changes in weather parameters across these locations, our system is able to enhance the spatial resolution of predictions and effectively detect anomalies in real-time. Additionally, our system employs unsupervised learning algorithms to identify unusual weather patterns, providing timely alerts. Our findings indicate that this system has the potential to enhance decision-making.

IKD+: Reliable Low Complexity Deep Models For Retinopathy Classification

Deep neural network (DNN) models for retinopathy have estimated predictive accuracies in the mid-to-high 90%. However, the following aspects remain unaddressed: State-of-the-art models are complex and require substantial computational infrastructure to train and deploy; The reliability of predictions can vary widely. In this paper, we focus on these aspects and propose a form of iterative knowledge distillation(IKD), called IKD+ that incorporates a tradeoff between size, accuracy and reliability. We investigate the functioning of IKD+ using two widely used techniques for estimating model calibration (Platt-scaling and temperature-scaling), using the best-performing model available, which is an ensemble of EfficientNets with approximately 100M parameters. We demonstrate that IKD+ equipped with temperature-scaling results in models that show up to approximately 500-fold decreases in the number of parameters than the original ensemble without a significant loss in accuracy. In addition, calibration scores (reliability) for the IKD+ models are as good as or better than the base mode