The moon is a treasure trove of valuable resources. Gold, platinum, and many rare Earth metals await extraction to be used in next-generation electronics. But there's one resource in particular that has excited scientists, rocket engineers, space agency officials, industry entrepreneurs--virtually anyone with a vested interest in making spaceflight to distant worlds more affordable. If you split water into hydrogen and oxygen, and then liquefy those constituents, you have rocket fuel. If you can stop at the moon's orbit or a lunar base to refuel, you no longer need to bring all your propellant with you as you take off, making your spacecraft significantly lighter and cheaper to launch.

Dimensionality reduction is the process of expressing high-dimensional data in a reduced number of dimensions such that each one contains the most amount of information. Dimensionality reduction may be used for visualization of high-dimensional data or to speed up machine learning models by removing low-information or correlated features. Principal Component Analysis, or PCA, is a popular method of reducing the dimensionality of data by drawing several orthogonal (perpendicular) vectors in the feature space to represent the reduced number of dimensions. The variable number represents the number of dimensions the reduced data will have. In the case of visualization, for example, it would be two dimensions.

Machine learning (ML) is the field of training machines to achieve high level of cognition and perform human-like analysis. Since ML is a data-driven approach, it seemingly fits into our daily lives and operations as well as complex and interdisciplinary fields. With the rise of commercial, open-source and user-catered ML tools, a key question often arises whenever ML is applied to explore a phenomenon or a scenario: what constitutes a good ML model? Keeping in mind that a proper answer to this question depends on a variety of factors, this work presumes that a good ML model is one that optimally performs and best describes the phenomenon on hand. From this perspective, identifying proper assessment metrics to evaluate performance of ML models is not only necessary but is also warranted. As such, this paper examines a number of the most commonly-used performance fitness and error metrics for regression and classification algorithms, with emphasis on engineering applications.

Their study, published today in Science Advances, is the first to apply an experimentally validated machine learning method to rapidly design and develop advanced gas separation membranes. "Our work points to a new way of materials design and we expect it to revolutionize the field," says the study's PI Sanat Kumar, Bykhovsky Professor of Chemical Engineering and a pioneer in developing polymer nanocomposites with improved properties. Plastic films or membranes are often used to separate mixtures of simple gases, like carbon dioxide (CO2), nitrogen (N2), and methane (CH4). Scientists have proposed using membrane technology to separate CO2 from other gases for natural gas purification and carbon capture, but there are potentially hundreds of thousands of plastics that can be produced with our current synthetic toolbox, all of which vary in their chemical structure. Manufacturing and testing all of these materials is an expensive and time-consuming process, and to date, only about 1,000 have been evaluated as gas separation membranes.

IMAGE: Researchers from U of T Engineering and Carnegie Mellon University are using electrolyzers like this one to convert waste CO2 into commercially valuable chemicals. Their latest catalyst, designed in part... view more Researchers at University of Toronto Engineering and Carnegie Mellon University are using artificial intelligence (AI) to accelerate progress in transforming waste carbon into a commercially valuable product with record efficiency. They leveraged AI to speed up the search for the key material in a new catalyst that converts carbon dioxide (CO2) into ethylene -- a chemical precursor to a wide range of products, from plastics to dish detergent. The resulting electrocatalyst is the most efficient in its class. If run using wind or solar power, the system also provides an efficient way to store electricity from these renewable but intermittent sources.

Researchers at University of Toronto Engineering and Carnegie Mellon University are using artificial intelligence (AI) to accelerate progress in transforming waste carbon into a commercially valuable product with record efficiency. They leveraged AI to speed up the search for the key material in a new catalyst that converts carbon dioxide (CO2) into ethylene--a chemical precursor to a wide range of products, from plastics to dish detergent. The resulting electrocatalyst is the most efficient in its class. If run using wind or solar power, the system also provides an efficient way to store electricity from these renewable but intermittent sources. "Using clean electricity to convert CO2 into ethylene, which has a $60 billion global market, can improve the economics of both carbon capture and clean energy storage," says Professor Ted Sargent, one of the senior authors on a new paper published today in Nature.

In this paper, we discuss the learning of generalised policies for probabilistic and classical planning problems using Action Schema Networks (ASNets). The ASNet is a neural network architecture that exploits the relational structure of (P)PDDL planning problems to learn a common set of weights that can be applied to any problem in a domain. By mimicking the actions chosen by a traditional, non-learning planner on a handful of small problems in a domain, ASNets are able to learn a generalised reactive policy that can quickly solve much larger instances from the domain. This work extends the ASNet architecture to make it more expressive, while still remaining invariant to a range of symmetries that exist in PPDDL problems. We also present a thorough experimental evaluation of ASNets, including a comparison with heuristic search planners on seven probabilistic and deterministic domains, an extended evaluation on over 18,000 Blocksworld instances, and an ablation study. Finally, we show that sparsity-inducing regularisation can produce ASNets that are compact enough for humans to understand, yielding insights into how the structure of ASNets allows them to generalise across a domain.

Ore deposits are generally formed by hydrothermal activity, which also forms various types of alteration zones in the vicinity of such deposits. Identifying alteration zones can clarify the mineralization mechanism and provides information indispensable for mineral resource exploration. The application of an alteration halo accompanied by the alteration of host rocks to the exploration of the Kuroko ore deposits produced many results in Japan (see, e.g., [1]). It is important to identify the alteration zone in the exploration of porphyry copper deposits. More than 50% of global copper production is from porphyry copper deposits, making them the most important copper resource.

The study of causal effects of an intervention or treatment on a specific outcome based on observational data is a fundamental problem in many applications. Examples include understanding the effects of massive wildfires on a person's mental health, of teaching methods on a student's employability, or of disease outbreaks on the global stock market. A critical problem of causal inference from observational data is confounding. A variable that affects both the treatment and the outcome is known as a confounder of the treatment effects on the outcome. Standard ways to measure observable confounders include propensity score matching and their variants (Rubin, 2005). However, if a confounder is hidden, the treatment effect on the outcome cannot be directly estimated without further assumptions (Pearl, 2009; Louizos et al., 2017). For example, household income, which cannot be easily measured, can affect both the therapy options available to a patient and the health condition after therapy of that patient.

Autonomous detection of desired events from large databases using time series classification is becoming increasingly important in civil engineering as a result of continued long-term health monitoring of a large number of engineering structures encompassing buildings, bridges, towers, and offshore platforms. In this context, this paper proposes the application of a relatively new time series representation named "Shapelet transform", which is based on local similarity in the shape of the time series subsequences. In consideration of the individual attributes distinctive to time series signals in earthquake, wind and ocean engineering, the application of this transform yields a new shape-based feature representation. Combining this shape-based representation with a standard machine learning algorithm, a truly "white-box" machine learning model is proposed with understandable features and a transparent algorithm. This model automates event detection without the intervention of domain practitioners, yielding a practical event detection procedure. The efficacy of this proposed shapelet transform-based autonomous detection procedure is demonstrated by examples, to identify known and unknown earthquake events from continuously recorded ground-motion measurements, to detect pulses in the velocity time history of ground motions to distinguish between near-field and far-field ground motions, to identify thunderstorms from continuous wind speed measurements, to detect large-amplitude wind-induced vibrations from the bridge monitoring data, and to identify plunging breaking waves that have a significant impact on offshore structures.