With the capability of accurately representing a functional relationship between the inputs of a physical system's model and output quantities of interest, neural networks have become popular for surrogate modeling in scientific applications. However, as these networks are over-parameterized, their training often requires a large amount of data. To prevent overfitting and improve generalization error, regularization based on, e.g., $\ell_1$- and $\ell_2$-norms of the parameters is applied. Similarly, multiple connections of the network may be pruned to increase sparsity in the network parameters. In this paper, we explore the effects of sparsity promoting $\ell_1$-regularization on training neural networks when only a small training dataset from a high-fidelity model is available. As opposed to standard $\ell_1$-regularization that is known to be inadequate, we consider two variants of $\ell_1$-regularization informed by the parameters of an identical network trained using data from lower-fidelity models of the problem at hand. These bi-fidelity strategies are generalizations of transfer learning of neural networks that uses the parameters learned from a large low-fidelity dataset to efficiently train networks for a small high-fidelity dataset. We also compare the bi-fidelity strategies with two $\ell_1$-regularization methods that only use the high-fidelity dataset. Three numerical examples for propagating uncertainty through physical systems are used to show that the proposed bi-fidelity $\ell_1$-regularization strategies produce errors that are one order of magnitude smaller than those of networks trained only using datasets from the high-fidelity models.

Goal-conditioned hierarchical reinforcement learning (HRL) serves as a successful approach to solving complex and temporally extended tasks. Recently, its success has been extended to more general settings by concurrently learning hierarchical policies and subgoal representations. However, online subgoal representation learning exacerbates the non-stationary issue of HRL and introduces challenges for exploration in high-level policy learning. In this paper, we propose a state-specific regularization that stabilizes subgoal embeddings in well-explored areas while allowing representation updates in less explored state regions. Benefiting from this stable representation, we design measures of novelty and potential for subgoals, and develop an efficient hierarchical exploration strategy that actively seeks out new promising subgoals and states. Experimental results show that our method significantly outperforms state-of-the-art baselines in continuous control tasks with sparse rewards and further demonstrate the stability and efficiency of the subgoal representation learning of this work, which promotes superior policy learning.

Quantum annealers, such as the device built by D-Wave Systems, Inc., offer a way to compute solutions of NP-hard problems that can be expressed in Ising or QUBO (quadratic unconstrained binary optimization) form. Although such solutions are typically of very high quality, problem instances are usually not solved to optimality due to imperfections of the current generations quantum annealers. In this contribution, we aim to understand some of the factors contributing to the hardness of a problem instance, and to use machine learning models to predict the accuracy of the D-Wave 2000Q annealer for solving specific problems. We focus on the Maximum Clique problem, a classic NP-hard problem with important applications in network analysis, bioinformatics, and computational chemistry. By training a machine learning classification model on basic problem characteristics such as the number of edges in the graph, or annealing parameters such as D-Wave's chain strength, we are able to rank certain features in the order of their contribution to the solution hardness, and present a simple decision tree which allows to predict whether a problem will be solvable to optimality with the D-Wave 2000Q. We extend these results by training a machine learning regression model that predicts the clique size found by D-Wave.

This paper discusses a new variant of the Henry Gas Solubility Optimization (HGSO) Algorithm, called Hybrid HGSO (HHGSO). Unlike its predecessor, HHGSO allows multiple clusters serving different individual meta-heuristic algorithms (i.e., with its own defined parameters and local best) to coexist within the same population. Exploiting the dynamic cluster-to-algorithm mapping via penalized and reward model with adaptive switching factor, HHGSO offers a novel approach for meta-heuristic hybridization consisting of Jaya Algorithm, Sooty Tern Optimization Algorithm, Butterfly Optimization Algorithm, and Owl Search Algorithm, respectively. The acquired results from the selected two case studies (i.e., involving team formation problem and combinatorial test suite generation) indicate that the hybridization has notably improved the performance of HGSO and gives superior performance against other competing meta-heuristic and hyper-heuristic algorithms.

The goal of Kaare Hartvig Jensen, Associate Professor at DTU Physics, was to reduce the need for harvesting, transporting, and processing crops for the production of biofuels, pharmaceuticals, and other products. The new method of extracting the necessary substances, which are called plant metabolites, also eliminates the need for chemical and mechanical processes. Plant metabolites consist of a wide range of extremely important chemicals. Many, such as the malaria drug artemisinin, have remarkable therapeutic properties, while others, like natural rubber or biofuel from tree sap, have mechanical properties. Because most plant metabolites are isolated in individual cells, the method of extracting the metabolites is also important, since the procedure affects both product purity and yield.

Engineering Application of Data Science can be defined as using Artificial Intelligence and Machine Learning to model physical phenomena purely based on facts (field measurements, data). The main objective of this technology is the complete avoidance of assumptions, simplifications, preconceived notions, and biases. One of the major characteristics of Engineering Application of Data Science is its incorporation of Explainable Artificial Intelligence (XAI). While using actual field measurements as the main building blocks of modeling physical phenomena, Engineering Application of Data Science incorporates several types of Machine Learning Algorithms including artificial neural networks, fuzzy set theory, and evolutionary computing. Predictive models of Engineering Application of Data Science (data-driven predictive models) are not represented through unexplainable "Black Box". Predictive models of Engineering Application of Data Science are reasonably explainable.

This summer, the whole world watched in horror as thousands of fires, again this year, ravaged the Amazon rainforest. Yet the forests are specific ecosystems: they are carbon sinks, meaning they stock carbon dioxide outside of the atmosphere; their destruction is contributing to climate change. To fight this phenomenon and protect the environment, governments, associations, scientists and local communities are relying more and more on technological advances. More specifically, here's how satellite imagery, artificial intelligence, and drones are being deployed in environmental battles. Combined with other sources of information (data collected in the field, aerial photography, etc.), satellite imagery makes it possible to analyse the evolution of forests, to detect changes that have arisen in a particular area and over a given period of time, and, ultimately, to determine the rate of global deforestation.

Daniel Faggella: So, Lorena, I want to be able to dive into these various use cases of how artificial intelligence can start to unlock the value of data in the oil and gas space, and make this really tangible. I know the first category we wanted to talk about was really around the value of subsurface data, that there's a lot of subsurface data, obviously in the oil and oil and gas domain. Lorena Pelegrín: And we see that AI or our ML can help these teams find the data and process the data much, much faster. Yeah, and I imagine a good deal of this has to do with, tell me if I'm wrong here, Lorena, but having an understanding of your company from working with you guys for a little while, I would imagine that the digitization of these myriad, somewhat chunky paper forms is one part of the process here, using some kind of optical character recognition stuff and working with historical records and maybe congealing and digitizing that. Daniel Faggella: But you let me know, Lorena, where does M&A, where does this data come in, in terms of the real value for potential M&A? Daniel Faggella: So Drone Deploy, for example, was on talking about what they do in the energy space with drones and video data to look at and inspect assets.

One of the most powerful supercomputers in the world has been brought online in the US and will be asked to apply its considerable artificial intelligence to some of the most challenging projects out there, from astrophysics and climate, to clean energy technologies. The Perlmutter system, a Hewlett-Packard-built Cray EX supercomputer, was unveiled at the National Energy Research Scientific Computing Center (NERSC) in California, part of the Lawrence Berkeley National Laboratory, and is "the fastest on the planet", according to Nvidia, the chip manufacturer supplying much of its graphics hardware. "The Perlmutter supercomputer will help inspire the next generation of scientists and innovators, allowing the US and Department of Energy to remain a leader in using scientific computation to answer our greatest questions," said David Turk, the US Department of Energy's deputy secretary, at its launch. "As we continue to enhance and deploy computing platforms like this, our national labs will only be better positioned to develop solutions to today's toughest problems, from climate change to cyber security." The machine is powered by 6,159 Nvidia A100 Tensor Core graphics processing units, rendering it capable of delivering almost four exaflops or a quintillion floating-point operations per second, according to Nvidia.

Optimism was the tale of the tape again today, amid news that half of the U.S. population is now fully vaccinated from COVID, with U.S. average daily cases falling below the 25,000-level. Bitcoin is also stabilizing and drove past the $40,000 mark again. Meanwhile, reopening plays moved the market, with stocks like Royal Caribbean gaining a whopping 11% week-to-date. Ford also rose more than 2% after pledging to increase its investment in EVs by $30 billion through 2025. This also comes on the heels of a landmark ruling by a Dutch court demanding oil giant Shell to cut carbon emissions by 45% no later than 2030.