Modeling strong gravitational lenses is computationally expensive for the complex data from modern and next-generation cosmic surveys. Deep learning has emerged as a promising approach for finding lenses and predicting lensing parameters, such as the Einstein radius. Mean-variance Estimators (MVEs) are a common approach for obtaining aleatoric (data) uncertainties from a neural network prediction. However, neural networks have not been demonstrated to perform well on out-of-domain target data successfully -- e.g., when trained on simulated data and applied to real, observational data. In this work, we perform the first study of the efficacy of MVEs in combination with unsupervised domain adaptation (UDA) on strong lensing data. The source domain data is noiseless, and the target domain data has noise mimicking modern cosmology surveys. We find that adding UDA to MVE increases the accuracy on the target data by a factor of about two over an MVE model without UDA. Including UDA also permits much more well-calibrated aleatoric uncertainty predictions. Advancements in this approach may enable future applications of MVE models to real observational data.

Modeling strong gravitational lenses is prohibitively expensive for modern and next-generation cosmic survey data. Neural posterior estimation (NPE), a simulation-based inference (SBI) approach, has been studied as an avenue for efficient analysis of strong lensing data. However, NPE has not been demonstrated to perform well on out-of-domain target data -- e.g., when trained on simulated data and then applied to real, observational data. In this work, we perform the first study of the efficacy of NPE in combination with unsupervised domain adaptation (UDA). The source domain is noiseless, and the target domain has noise mimicking modern cosmology surveys. We find that combining UDA and NPE improves the accuracy of the inference by 1-2 orders of magnitude and significantly improves the posterior coverage over an NPE model without UDA. We anticipate that this combination of approaches will help enable future applications of NPE models to real observational data.

Brian Nord first fell in love with physics when he was a teenager growing up in Wisconsin. His high school physics program wasn't exceptional, and he sometimes struggled to keep up with class material, but those difficulties did nothing to dampen his interest in the subject. In addition to the main curriculum, students were encouraged to independently study topics they found interesting, and Nord quickly developed a fascination with the cosmos. "A touchstone that I often come back to is space," he says. Nord was an avid reader of comic books, and astrophysics appealed to his desire to become a part of something bigger.

Physicists are increasingly developing artificial intelligence and machine learning techniques to advance our understanding of the physical world but there is a rising concern about the bias in such systems and their wider impact on society at large. In 2011, during her undergraduate degree at Georgia Institute of Technology, Ghanaian-US computer scientist Joy Buolamwini discovered that getting a robot to play a simple game of peek-a-boo with her was impossible – the machine was incapable of seeing her dark-skinned face. Later, in 2015, as a Master's student at Massachusetts Institute of Technology's Media Lab working on a science–art project called Aspire Mirror, she had a similar issue with facial analysis software: it detected her face only when she wore a white mask. Buolamwini's curiosity led her to run one of her profile images across four facial recognition demos, which, she discovered, either couldn't identify a face at all or misgendered her – a bias that she refers to as the "coded gaze". She then decided to test 1270 faces of politicians from three African and three European countries, with different features, skin tones and gender, which became her Master's thesis project "Gender Shades: Intersectional accuracy disparities in commercial gender classification" (figure 1).

Physicists built the Large Hadron Collider to study the inner workings of the universe. Inside a 27-kilometer underground ring straddling the French-Swiss border, the machine smashes protons together at nearly the speed of light to produce--fleetingly--the smallest constituent building blocks of nature. Sifting through snapshots of these collisions, LHC researchers look for new particles and scrutinize known ones, including their most famous find, in 2012: the Higgs boson, whose behavior explains why other fundamental particles like electrons and quarks have mass. Less well known is the intricate software engine that powers such discoveries. With particle collisions occurring at approximately a billion times per second, the facility generates about 40 terabytes of data per second, according to LHC physicist Maurizio Pierini.

Cluster algorithms are gaining in popularity due to their compelling ability to identify discrete subgroups in data, and their increasing accessibility in mainstream programming languages and statistical software. While researchers can follow guidelines to choose the right algorithms, and to determine what constitutes convincing clustering, there are no firmly established ways of computing a priori statistical power for cluster analysis. Here, we take a simulation approach to estimate power and classification accuracy for popular analysis pipelines. We systematically varied cluster size, number of clusters, number of different features between clusters, effect size within each different feature, and cluster covariance structure in generated datasets. We then subjected these datasets to common dimensionality reduction approaches (none, multi-dimensional scaling, or uniform manifold approximation and projection) and cluster algorithms (k-means, hierarchical agglomerative clustering with Ward linkage and Euclidean distance, or average linkage and cosine distance, HDBSCAN). Furthermore, we simulated additional datasets to explore the effect of sample size and cluster separation on statistical power and classification accuracy. We found that clustering outcomes were driven by large effect sizes or the accumulation of many smaller effects across features, and were mostly unaffected by differences in covariance structure. Sufficient statistical power can be achieved with relatively small samples (N=20 per subgroup), provided cluster separation is large ({\Delta}=4). Finally, we discuss whether fuzzy clustering (c-means) could provide a more parsimonious alternative for identifying separable multivariate normal distributions, particularly those with lower centroid separation.

Brian Nord wants to build a self-driving telescope. The Fermilab astrophysicist envisions an instrument that, when presented with a hypothesis about the nature of the Universe, figures out the best observations to make on its own. He anticipates that it could take up to thirty years to understand and put together the project's nuts and bolts. One known component is artificial intelligence (AI)--algorithms similar to those that underpin facial recognition software and nascent self-driving car technology. Building toward his telescope dream, Nord has begun applying AI to problems in astronomy, such as identifying unusual astronomical objects known as gravitational lenses.

NORD (Neural Operations Research & Development) is an open source distributed deep learning architectural research framework, based on PyTorch, MPI and Horovod. It aims to make research of deep architectures easier for experts of different domains, in order to accelerate the process of finding better architectures, as well as study the best architectures generated for different datasets. Although currently under heavy development, the framework aims to allow the easy implementation of different design and optimization method families (optimization algorithms, meta-heuristics, reinforcement learning etc.) as well as the fair comparison between them. Furthermore, due to the computational resources required in order to optimize and evaluate network architectures, it leverage the use of distributed computing, while aiming to minimize the researcher's overhead required to implement it. Moreover, it strives to make the creation of architectures more intuitive, by implementing network descriptors, allowing to separately define the architecture's nodes and connections. In this paper, we present the framework's current state of development, while presenting its basic concepts, providing simple examples as well as their experimental results.

Google uses "deep learning" to generate captions for images, Facebook uses it to recognize faces and Tesla uses it to train self-driving cars. Now astronomers have caught on to deep learning, a form of machine learning in which a computer can be trained to identify or classify particular objects in images. The newest telescopes, such as the Dark Energy Survey, which uses a 4-meter telescope in northern Chile and covers about one quarter of the southern sky, take millions of images of a variety of celestial objects. These often include visual distortions, cosmic rays and satellite trails that make them difficult to interpret. Deep learning could help process this deluge of data quickly.