Multi-stage stochastic optimization (MSSO) considers the problem of optimizing a sequence of decisions over a finite number of stages in the presence of stochastic observations, minimizing an expected cost while ensuring stage-wise action constraints are satisfied (Birge and Louveaux, 2011; Shapiro et al., 2014). Such a problem formulation captures a diversity of real-world process optimization problems, such as asset allocation (Dantzig and Infanger, 1993), inventory control (Shapiro et al., 2014; Nambiar et al., 2021), energy planning (Pereira and Pinto, 1991), and bio-chemical process control (Bao et al., 2019), to name a few. Despite the importance and ubiquity of the problem, it has proved challenging to develop algorithms that can cope with high-dimensional action spaces and long-horizon problems (Shapiro and Nemirovski, 2005; Shapiro, 2006). There have been a number of attempts to design scalable algorithms for MSSO, which generally attempt to exploit scenarios-wise or stage-wise decompositions. An example of a scenario-wise approach is Rockafellar and Wets (1991), which proposed a progressive hedging algorithm that decomposes the sample averaged approximation of the problem into individual scenarios and applies an augmented Lagrangian method to achieve consistency in a final solution.

Electron density $\rho(\vec{r})$ is the fundamental variable in the calculation of ground state energy with density functional theory (DFT). Beyond total energy, features in $\rho(\vec{r})$ distribution and modifications in $\rho(\vec{r})$ are often used to capture critical physicochemical phenomena in functional materials and molecules at the electronic scale. Methods providing access to $\rho(\vec{r})$ of complex disordered systems with little computational cost can be a game changer in the expedited exploration of materials phase space towards the inverse design of new materials with better functionalities. We present a machine learning framework for the prediction of $\rho(\vec{r})$. The model is based on equivariant graph neural networks and the electron density is predicted at special query point vertices that are part of the message passing graph, but only receive messages. The model is tested across multiple data sets of molecules (QM9), liquid ethylene carbonate electrolyte (EC) and LixNiyMnzCo(1-y-z)O2 lithium ion battery cathodes (NMC). For QM9 molecules, the accuracy of the proposed model exceeds typical variability in $\rho(\vec{r})$ obtained from DFT done with different exchange-correlation functional and show beyond the state of the art accuracy. The accuracy is even better for the mixed oxide (NMC) and electrolyte (EC) datasets. The linear scaling model's capacity to probe thousands of points simultaneously permits calculation of $\rho(\vec{r})$ for large complex systems many orders of magnitude faster than DFT allowing screening of disordered functional materials.

Sampling random variables following a Boltzmann distribution is an NP-hard problem involved in various applications such as training of \textit{Boltzmann machines}, a specific kind of neural network. Several attempts have been made to use a D-Wave quantum computer to sample such a distribution, as this could lead to significant speedup in these applications. Yet, at present, several challenges remain to efficiently perform such sampling. We detail the various obstacles and explain the remaining difficulties in solving the sampling problem on a D-wave machine.

The continuous electrification of the mobility and heating sectors adds much-needed flexibility to the power system. However, flexibility utilization also introduces new challenges to distribution system operators (DSOs), who need mechanisms to supervise flexibility activations and monitor their effect on distribution network operation. Flexibility activations can be broadly categorized to those originating from electricity markets and those initiated by the DSO to avoid constraint violations. Simultaneous electricity market driven flexibility activations may cause voltage quality or temporary overloading issues, and the failure of flexibility activations initiated by the DSO might leave critical grid states unresolved. This work proposes a novel data processing pipeline for automated real-time identification of fast-ramped flexibility activation events. Its practical value is twofold: i) potentially critical flexibility activations originating from electricity markets can be detected by the DSO at an early stage, and ii) successful activation of DSO-requested flexibility can be verified by the operator. In both cases the increased awareness would allow the DSO to take counteractions to avoid potentially critical grid situations. The proposed pipeline combines techniques from unsupervised detection and open-set classification. For both building blocks feasibility is systematically evaluated and proofed on real load and flexibility activation data.

We consider the problem of learning an episodic safe control policy that minimizes an objective function, while satisfying necessary safety constraints -- both during learning and deployment. We formulate this safety constrained reinforcement learning (RL) problem using the framework of a finite-horizon Constrained Markov Decision Process (CMDP) with an unknown transition probability function. Here, we model the safety requirements as constraints on the expected cumulative costs that must be satisfied during all episodes of learning. We propose a model-based safe RL algorithm that we call the Optimistic-Pessimistic Safe Reinforcement Learning (OPSRL) algorithm, and show that it achieves an $\tilde{\mathcal{O}}(S^{2}\sqrt{A H^{7}K}/ (\bar{C} - \bar{C}_{b}))$ cumulative regret without violating the safety constraints during learning, where $S$ is the number of states, $A$ is the number of actions, $H$ is the horizon length, $K$ is the number of learning episodes, and $(\bar{C} - \bar{C}_{b})$ is the safety gap, i.e., the difference between the constraint value and the cost of a known safe baseline policy. The scaling as $\tilde{\mathcal{O}}(\sqrt{K})$ is the same as the traditional approach where constraints may be violated during learning, which means that our algorithm suffers no additional regret in spite of providing a safety guarantee. Our key idea is to use an optimistic exploration approach with pessimistic constraint enforcement for learning the policy. This approach simultaneously incentivizes the exploration of unknown states while imposing a penalty for visiting states that are likely to cause violation of safety constraints. We validate our algorithm by evaluating its performance on benchmark problems against conventional approaches.

An outlier is an event or observation that is defined as an unusual activity, intrusion, or a suspicious data point that lies at an irregular distance from a population. The definition of an outlier event, however, is subjective and depends on the application and the domain (Energy, Health, Wireless Network, etc.). It is important to detect outlier events as carefully as possible to avoid infrastructure failures because anomalous events can cause minor to severe damage to infrastructure. For instance, an attack on a cyber-physical system such as a microgrid may initiate voltage or frequency instability, thereby damaging a smart inverter which involves very expensive repairing. Unusual activities in microgrids can be mechanical faults, behavior changes in the system, human or instrument errors or a malicious attack. Accordingly, and due to its variability, Outlier Detection (OD) is an ever-growing research field. In this chapter, we discuss the progress of OD methods using AI techniques. For that, the fundamental concepts of each OD model are introduced via multiple categories. Broad range of OD methods are categorized into six major categories: Statistical-based, Distance-based, Density-based, Clustering-based, Learning-based, and Ensemble methods. For every category, we discuss recent state-of-the-art approaches, their application areas, and performances. After that, a brief discussion regarding the advantages, disadvantages, and challenges of each technique is provided with recommendations on future research directions. This survey aims to guide the reader to better understand recent progress of OD methods for the assurance of AI.

Hadi Ganjineh is Head of IT, Integrated Tech & Innovation at Super Energy Corp. focused on leading-edge digital business, tech & innovation. We use energy in many different ways in our lives, be it for lighting up our houses, running electronic appliances or as fuel in our vehicles. There are mainly two types of energy: renewable energy and non-renewable energy. Non-renewable energy includes fossil fuels like natural gas, petroleum and coal. However, these energy sources come from nature itself; it is impossible to renew them quickly.

Hongliang Xin, an associate professor of chemical engineering in the College of Engineering, and his collaborators have devised a new artificial intelligence framework that can accelerate discovery of materials for important technologies, such as fuel cells and carbon capture devices. Titled "Infusing theory into deep learning for interpretable reactivity prediction," their paper in the journal Nature Communications details a new approach called TinNet -- short for theory-infused neural network -- that combines machine-learning algorithms and theories for identifying new catalysts. Catalysts are materials that trigger or speed up chemical reactions. TinNet is based on deep learning, also known as a subfield of machine learning, which uses algorithms to mimic how human brains work. The 1996 victory of IBM's Deep Blue computer over world chess champion Garry Kasparov was one of the first advances in machine learning.

Artificial intelligence (AI) offers enormous potential to accelerate technological development in nuclear fields, from science to energy to medicine, and the sector is making good progress in seizing on those opportunities, according to speakers in webinars organized by the International Telecommunication Union (ITU) in partnership with the IAEA. AI for Nuclear Energy, held on 24 November and attracting more than 1200 participants, was one the most popular sessions of ITU's AI for Good Global Summit 2021. It showcased efforts to capitalize on technological advancements in artificial intelligence to enhance the development and deployment of nuclear power, enabling this low-carbon energy source to fulfil its potential in the fight against climate change and meeting the goals of the 2030 Agenda and the Paris Agreement. "In order to be competitive, as well as integrated into the mix of modern energy systems, nuclear power plants – in addition to being safe, secure and reliable – also need to be economical and efficient," said Mikhail Chudakov, IAEA Deputy Director General and Head of Department of Nuclear Energy, in his welcome remarks. "AI-based approaches can contribute to these areas."

The energy grid is a complex network of hard and soft infrastructure that delivers electricity from producers to consumers. Producing the electricity that powers our homes and businesses involves dozens of steps, including generation, transmission, distribution, and consumption. Luckily, most people in the United States don't have to think about this process. They simply pay the electricity bill each month and the lights come on. The electricity grid in the United States has remained relatively stagnant for decades.