Search

The increasing intensity and frequency of floods is one of the many consequences of our changing climate. In this work, we explore ML techniques that improve the flood detection module of an operational early flood warning system. Our method exploits an unlabelled dataset of paired multi-spectral and Synthetic Aperture Radar (SAR) imagery to reduce the labeling requirements of a purely supervised learning method. Prior works have used unlabelled data by creating weak labels out of them. However, from our experiments we noticed that such a model still ends up learning the label mistakes in those weak labels. Motivated by knowledge distillation and semi supervised learning, we explore the use of a teacher to train a student with the help of a small hand labelled dataset and a large unlabelled dataset. Unlike the conventional self distillation setup, we propose a cross modal distillation framework that transfers supervision from a teacher trained on richer modality (multi-spectral images) to a student model trained on SAR imagery. The trained models are then tested on the Sen1Floods11 dataset. Our model outperforms the Sen1Floods11 baseline model trained on the weak labeled SAR imagery by an absolute margin of 6.53% Intersection-over-Union (IoU) on the test split.

The operational flood forecasting system by Google was developed to provide accurate real-time flood warnings to agencies and the public, with a focus on riverine floods in large, gauged rivers. It became operational in 2018 and has since expanded geographically. This forecasting system consists of four subsystems: data validation, stage forecasting, inundation modeling, and alert distribution. Machine learning is used for two of the subsystems. Stage forecasting is modeled with the Long Short-Term Memory (LSTM) networks and the Linear models. Flood inundation is computed with the Thresholding and the Manifold models, where the former computes inundation extent and the latter computes both inundation extent and depth. The Manifold model, presented here for the first time, provides a machine-learning alternative to hydraulic modeling of flood inundation. When evaluated on historical data, all models achieve sufficiently high-performance metrics for operational use. The LSTM showed higher skills than the Linear model, while the Thresholding and Manifold models achieved similar performance metrics for modeling inundation extent. During the 2021 monsoon season, the flood warning system was operational in India and Bangladesh, covering flood-prone regions around rivers with a total area of 287,000 km2, home to more than 350M people. More than 100M flood alerts were sent to affected populations, to relevant authorities, and to emergency organizations. Current and future work on the system includes extending coverage to additional flood-prone locations, as well as improving modeling capabilities and accuracy., Comment: 36 pages, 10 figures, 3 tables, 1 supplementary table (9 pages)

An unconventional solution for finding the location of event creation is presented. It is based on two feed-forward neural networks with fixed architecture, whose parameters are chosen so as to reach a high accuracy. The interaction point location is a parameter that can be used to select events of interest from the very high rate of events created at the current experiments in High Energy Physics. The system suggested here is tested on simulated data sets of the ZEUS Central Tracking Detector, and is shown to perform better than conventional algorithms., Comment: Presentes at ACAT 2000, FermiLab, Chicago, October 2000

A neural network solution for a complicated experimental High Energy Physics problem is described. The method is used to reconstruct the momentum and charge of muons produced in collisions of particle in the ATLAS detector. The information used for the reconstruction is limited to the output of the outer layer of the detector, after the muons went through strong and inhomogeneous magnetic field that have bent their trajectory. It is demonstrated that neural network solution is efficient in performing this task. It is shown that this mechanism can be efficient in rapid classification as required in triggering systems of the future particle accelerators. The parallel processing nature of the network makes it relevant for hardware realization in the ATLAS triggering system., Comment: Presented at ACAT 2000, FermiLab, Chicago, October 2000

In July 1969, an Israeli parliament decision instigated an educational reform focused on curriculum development, teacher education, new buildings, new equipment, and a change from 8 to 6 years in elementary school and from 4 to 6 years in secondary school. Dropouts, useless curricula, job market demands for better education, and integration and academic achievement needs necessitated the reform. Elements of the decision included implementation, teacher training, the future educational structure, establishing an advisory committee, and building junior high schools. A central committee, a subcommittee for administrative matters, and a subcommittee for teaching issues constituted the structure to execute the change. Teacher resistance, parent resistance, and financial restrictions obstructed the government's objectives. Among the ministry's strategies were public committees to obtain public support and the introduction of normative value-based and broad social goals. The first major objective--to raise the level of studies--was to be accomplished through counseling; attention to underprivileged students; and changes in curriculum, teacher training, teaching methods, and school structure. The second objective--integration--was to be achieved through new zoning and the eradication of the selection process. Results of the changes include increased enrollment, fewer dropouts, and byproducts such as discussions concerning Israeli society and the quality of education. (RG)