A way of directly measuring the carbon dioxide released by burning fossil fuels could help cities and countries monitor their efforts to reduce emissions in near real time. "We are in a shrinking window of time to do this, so I think we really need to know what the situation is as quickly and as accurately as possible," says Penelope Pickers at the University of East Anglia, UK. At present, governments and research organisations estimate countries' overall emissions based on data such as how much oil or gas has been sold. While initial estimates are often made fairly quickly, it can take years to fully compile this information and estimates can vary substantially. Measuring fossil fuel emissions directly would help confirm the accuracy of these inventory-based estimates and reveal more quickly if emission-reduction policies are working or not.

Machine learning models help with battery life cycle management. Blending advanced electronics with IoT, data science and digital twins, ML models with predictive intelligence can anticipate battery life, identify degradation breakdowns and their causes. Akhil Aryan, the cofounder of ION Energy, a company that applies intelligent battery analytics to improve the performance of lithium-ion batteries, says data on battery life includes performance, state of charge, stress from rapid acceleration and deceleration, temperature and the number of charge cycles.

Dutch submersible manufacturer U-Boat Worx breaks the mould with the launch of the NEXUS series. U-Boat Worx is the market leader in private and commercial submersibles. Since 2017, the company has sold more than 20 of its highly successful Cruise Subs to private operators, resorts, and cruise lines. The NEXUS series comprises two models featuring an ultra-large elliptical acrylic pressure hull with unrivalled passenger comfort. Seating up to eight passengers and one pilot, the NEXUS provides 25% more interior space than competing models.

The energy transition introduces more volatile energy sources into the power grids. In this context, power transfer between different synchronous areas through High Voltage Direct Current (HVDC) links becomes increasingly important. Such links can balance volatile generation by enabling long-distance transport or by leveraging their fast control behavior. Here, we investigate the interaction of power imbalances - represented through the power grid frequency - and power flows on HVDC links between synchronous areas in Europe. We use explainable machine learning to identify key dependencies and disentangle the interaction of critical features. Our results show that market-based HVDC flows introduce deterministic frequency deviations, which however can be mitigated through strict ramping limits. Moreover, varying HVDC operation modes strongly affect the interaction with the grid. In particular, we show that load-frequency control via HVDC links can both have control-like or disturbance-like impacts on frequency stability.

This blog is accompanied by a Colab notebook which provides an in-depth look at how Raster Vision works and allows you to run each experiment discussed in this post yourself. Change detection is the computer-vision equivalent of the spot-the-difference game. Given two images, the model must detect all the points at which they differ. In the context of remote sensing, these images are usually satellite or aerial images of the same geographical location at two different points in time. Change detection has been an active research area for a long time and the literature is rich with algorithms that perform the task automatically, ranging from basic image processing techniques to present-day deep neural networks.

Slowing down climate change is an urgent matter. If we fail, our world will face a more extensive crisis than we experienced because of the global COVID-19 pandemic. When artificial intelligence (AI) technology helps solve a problem, problem-solving can be done quicker, and the solution is often one that would have taken longer for humans to discover. There's no time to waste: atmospheric CO2 levels are the highest ever (even with significant drops from the stay-at-home orders for COVID-19), average sea levels are rising (3 inches in the last 25 years alone), and 2019 was the hottest year on record for the world's oceans. Artificial intelligence isn't a silver bullet, but it can certainly help us reduce greenhouse gas (GHG) emissions in various ways.

Abstract: The spread of COVID-19 has coincided with the rise of Graph Neural Networks (GNNs), leading to several studies proposing their use to better forecast the evolution of the pandemic. Many such models also include Long Short TermMemory (LSTM) networks, a common tool for time series forecasting. In this work, we further investigate the integration of these two methods by implementing GNNs within the gates of an LSTM and exploiting spatial information. In addition, we introduce a skip connection which proves critical to jointly capture the spatial and temporal patterns in the data. We validate our daily COVID-19 new cases forecast model on data of 37 European nations for the last 472 days and show superior performance compared to state-of-the-art graph time series models based on mean absolute scaled error (MASE).

At RapidMiner, we all know the power that artificial intelligence has to positively shape the future--in business and in the world at large. While many enterprises are relatively new to implementing AI, I've spent years scrutinizing prominent uses cases and staying on top of the latest trends. The hype around AI has only grown in 2022, but sustaining trends are what differentiates baseless hype from cold, hard reality. In this post, I'll break down the hype and the buzzwords to walk through what I consider to be the top five current AI and machine learning trends and how I predict they'll impact data science in the years to come. As I laid out in my data science manifesto, accountability is an essential part of a data scientist's role.

Recurrent neural networks have seen widespread use in modeling dynamical systems in varied domains such as weather prediction, text prediction and several others. Often one wishes to supplement the experimentally observed dynamics with prior knowledge or intuition about the system. While the recurrent nature of these networks allows them to model arbitrarily long memories in the time series used in training, it makes it harder to impose prior knowledge or intuition through generic constraints. In this work, we present a path sampling approach based on principle of Maximum Caliber that allows us to include generic thermodynamic or kinetic constraints into recurrent neural networks. We show the method here for a widely used type of recurrent neural network known as long short-term memory network in the context of supplementing time series collected from different application domains. These include classical Molecular Dynamics of a protein and Monte Carlo simulations of an open quantum system continuously losing photons to the environment and displaying Rabi oscillations. Our method can be easily generalized to other generative artificial intelligence models and to generic time series in different areas of physical and social sciences, where one wishes to supplement limited data with intuition or theory based corrections.

The practical importance of coherent forecasts in hierarchical forecasting has inspired many studies on forecast reconciliation. Under this approach, so-called base forecasts are produced for every series in the hierarchy and are subsequently adjusted to be coherent in a second reconciliation step. Reconciliation methods have been shown to improve forecast accuracy, but will, in general, adjust the base forecast of every series. However, in an operational context, it is sometimes necessary or beneficial to keep forecasts of some variables unchanged after forecast reconciliation. In this paper, we formulate reconciliation methodology that keeps forecasts of a pre-specified subset of variables unchanged or "immutable". In contrast to existing approaches, these immutable forecasts need not all come from the same level of a hierarchy, and our method can also be applied to grouped hierarchies. We prove that our approach preserves unbiasedness in base forecasts. Our method can also account for correlations between base forecasting errors and ensure non-negativity of forecasts. We also perform empirical experiments, including an application to sales of a large scale online retailer, to assess the impacts of our proposed methodology.