Time series data analysis is a critical component in various domains such as finance, healthcare, and meteorology. Despite the progress in deep learning for time series analysis, there remains a challenge in addressing the non-stationary nature of time series data. Traditional models, which are built on the assumption of constant statistical properties over time, often struggle to capture the temporal dynamics in realistic time series, resulting in bias and error in time series analysis. This paper introduces the Adaptive Wavelet Network (AdaWaveNet), a novel approach that employs Adaptive Wavelet Transformation for multi-scale analysis of non-stationary time series data. AdaWaveNet designed a lifting scheme-based wavelet decomposition and construction mechanism for adaptive and learnable wavelet transforms, which offers enhanced flexibility and robustness in analysis. We conduct extensive experiments on 10 datasets across 3 different tasks, including forecasting, imputation, and a newly established super-resolution task. The evaluations demonstrate the effectiveness of AdaWaveNet over existing methods in all three tasks, which illustrates its potential in various real-world applications.

Active learning optimizes the exploration of large parameter spaces by strategically selecting which experiments or simulations to conduct, thus reducing resource consumption and potentially accelerating scientific discovery. A key component of this approach is a probabilistic surrogate model, typically a Gaussian Process (GP), which approximates an unknown functional relationship between control parameters and a target property. However, conventional GPs often struggle when applied to systems with discontinuities and non-stationarities, prompting the exploration of alternative models. This limitation becomes particularly relevant in physical science problems, which are often characterized by abrupt transitions between different system states and rapid changes in physical property behavior. Fully Bayesian Neural Networks (FBNNs) serve as a promising substitute, treating all neural network weights probabilistically and leveraging advanced Markov Chain Monte Carlo techniques for direct sampling from the posterior distribution. This approach enables FBNNs to provide reliable predictive distributions, crucial for making informed decisions under uncertainty in the active learning setting. Although traditionally considered too computationally expensive for 'big data' applications, many physical sciences problems involve small amounts of data in relatively low-dimensional parameter spaces. Here, we assess the suitability and performance of FBNNs with the No-U-Turn Sampler for active learning tasks in the 'small data' regime, highlighting their potential to enhance predictive accuracy and reliability on test functions relevant to problems in physical sciences.

Transformers are one of the key assets in AC distribution grids and renewable power integration. During transformer energization inrush currents appear, which lead to transformer degradation and can cause grid instability events. These inrush currents are a consequence of the transformer's magnetic core saturation during its connection to the grid. Transformer cores are normally modelled by the Jiles-Atherton (JA) model which contains five parameters. These parameters can be estimated by metaheuristic-based search algorithms. The parameter initialization of these algorithms plays an important role in the algorithm convergence. The most popular strategy used for JA parameter initialization is a random uniform distribution. However, techniques such as parameter initialization by Probability Density Functions (PDFs) have shown to improve accuracy over random methods. In this context, this research work presents a framework to assess the impact of different parameter initialization strategies on the performance of the JA parameter estimation for inrush current studies. Depending on available data and expert knowledge, uncertainty levels are modelled with different PDFs. Moreover, three different metaheuristic-search algorithms are employed on two different core materials and their accuracy and computational time are compared. Results show an improvement in the accuracy and computational time of the metaheuristic-based algorithms when PDF parameter initialization is used.

Energy storage devices, such as batteries, thermal energy storages, and hydrogen systems, can help mitigate climate change by ensuring a more stable and sustainable power supply. To maximize the effectiveness of such energy storage, determining the appropriate charging and discharging amounts for each time period is crucial. Reinforcement learning is preferred over traditional optimization for the control of energy storage due to its ability to adapt to dynamic and complex environments. However, the continuous nature of charging and discharging levels in energy storage poses limitations for discrete reinforcement learning, and time-varying feasible charge-discharge range based on state of charge (SoC) variability also limits the conventional continuous reinforcement learning. In this paper, we propose a continuous reinforcement learning approach that takes into account the time-varying feasible charge-discharge range. An additional objective function was introduced for learning the feasible action range for each time period, supplementing the objectives of training the actor for policy learning and the critic for value learning. This actively promotes the utilization of energy storage by preventing them from getting stuck in suboptimal states, such as continuous full charging or discharging. This is achieved through the enforcement of the charging and discharging levels into the feasible action range. The experimental results demonstrated that the proposed method further maximized the effectiveness of energy storage by actively enhancing its utilization.

One can improve predictability in the unknown domain by combining forecasts of imperfect complex computational models using a Bayesian statistical machine learning framework. In many cases, however, the models used in the mixing process are similar. In addition to contaminating the model space, the existence of such similar, or even redundant, models during the multimodeling process can result in misinterpretation of results and deterioration of predictive performance. In this work we describe a method based on the Principal Component Analysis that eliminates model redundancy. We show that by adding model orthogonalization to the proposed Bayesian Model Combination framework, one can arrive at better prediction accuracy and reach excellent uncertainty quantification performance.

A survey of five million distant solar systems, aided by'neural network' algorithms, has discovered 60 stars that appear to be surrounded by giant alien power plants. Seven of the stars -- so-called M-dwarf stars that range between 60 percent and 8 percent the size of our sun -- were recorded giving off unexpectedly high infrared'heat signatures,' according to the astronomers. Natural, and better understood, outer space'phenomena,' as they report in their new study, 'cannot easily account for the observed infrared excess emission.' Ever since theoretical physicist Freeman Dyson first proposed the idea at Princeton in 1960, astrophysicists have speculated that advanced extraterrestrials might have constructed massive solar energy collectors around one star or more. While powering their spacefaring ET civilizations, these hypothetical'Dyson spheres,' would reveal themselves by radiating more heat than usual, the physicist argued. A survey of five million distant solar systems, aided by'neural networks,' has discovered 60 stars that appear to be surrounded by gigantic alien power plants.

Silicon Valley wants to use nuclear power to support the energy-hungry data centres that help train and deploy its artificial intelligence models. But realistic timelines show that any US nuclear renaissance will have at best a limited impact during a period of fast-rising electricity demand. Global electricity usage from data centres is already on track to double by 2026. In the US, data centres represent the fastest-growing source of energy demand at a time when the country's…

Technology giant Microsoft recently released its sustainability report for the 2023 financial year, and it didn't exactly have positive numbers. Microsoft set a climate goal in 2020 to become carbon negative by 2030, sequestering more carbon dioxide from the atmosphere than it emits, but the company seems to be on the wrong track to achieve this goal. Microsoft's greenhouse gas emissions increased by 30 percent in the 2023 financial year -- and it's all Copilot's fault. The big culprit is the company's huge AI investments. It takes huge amounts of energy to train and use AI models.

When Microsoft pledged four years ago to remove more carbon than it emits by the end of the decade, it was one of the most ambitious and comprehensive plans to tackle climate change. Now the software giant's relentless push to be the global leader in artificial intelligence is putting that goal in peril. The Seattle-based company's total planet-warming impact is about 30% higher today than it was in 2020, according to the latest sustainability report published Wednesday. That makes getting to below zero by 2030 even harder than it was when it announced its carbon-negative goal. Now to meet its goals, the software giant will have to make serious progress very quickly in gaining access to green steel and concrete and less carbon-intensive chips, said Brad Smith, president of Microsoft, in an exclusive interview with Bloomberg Green.

Quadrotors are widely employed across various domains, yet the conventional type faces limitations due to underactuation, where attitude control is closely tied to positional adjustments. In contrast, quadrotors equipped with tiltable rotors offer overactuation, empowering them to track both position and attitude trajectories. However, the nonlinear dynamics of the drone body and the sluggish response of tilting servos pose challenges for conventional cascade controllers. In this study, we propose a control methodology for tilting-rotor quadrotors based on nonlinear model predictive control (NMPC). Unlike conventional approaches, our method preserves the full dynamics without simplification and utilizes actuator commands directly as control inputs. Notably, we incorporate a first-order servo model within the NMPC framework. Through simulation, we observe that integrating the servo dynamics not only enhances control performance but also accelerates convergence. To assess the efficacy of our approach, we fabricate a tiltable-quadrotor and deploy the algorithm onboard at a frequency of 100Hz. Extensive real-world experiments demonstrate rapid, robust, and smooth pose tracking performance.