We are requesting a DSI Scholar position for an undergraduate to work with myself and my collaborator Ben Holtzman (Lamont Doherty Earth Observatory, LDEO). We have been collaborating on the development of novel machine learning applications to seismology, specifically unsupervised feature extraction in spectral properties of large numbers of small earthquakes. Our first application was published in Science Advances last year (Holtzman, Pate, Paisley, Waldhauser, Repetto, “Machine learning reveals cyclic changes in seismic source spectra in Geysers geothermal field.” Science Advances 4, eaao2929. doi:10.1126/sciadv.aao2929, 2018). Currently we are building a synthetic dataset to better understand the features that control clustering behavior, and compare different clustering methods.

Continue reading

Recently, there have been multiple failures of large tailings dams that store mining wastes, around the world, with devastating impacts (e.g., https://en.wikipedia.org/wiki/Brumadinho_dam_disaster). These dams are unique in that they continue to be raised as waste piles up and can get as tall as 400m. The risk and impact of failure increases as the dam gets taller. There are several thousand such dams around the world. The concept of the project is to develop a continuous status monitoring and risk analysis of these dams, automatically, using globally available satellite data from multiple bands, as well as regularly updated climate data products. Overtopping of the dam during an intense or persistent rainfall event is the leading mode of failure. Foundation failure which leads to a liquefaction or deformation of the dam is the second leading failure mode.

Continue reading

Galaxies in our universe form hierarchically, continuously merging and absorbing smaller galaxies over cosmic time. In this project we aim to identify the most important features of, as well as generate efficient new features from, the merger histories of galaxies. Namely, features that predict (or physically speaking, determine) the properties of galaxies, e.g. their shape or color. This will be done using the results from a large cosmological simulation, IllustrisTNG (www.tng-project.org). We will begin with identifying ways to represent the rich information in the merger history. We will then compare various ML methods oriented towards feature selection or importance analysis: random forests or gradient boosted trees, L1SVM, neural networks (through analysis of e.g. saliency maps). More advanced models can also be applied, such as neural network models designed for feature selection. Finally, we wish to apply / develop methods that can build ‘interpretable’ new features by constructing them as algebraic formulas from original input features (inspired by e.g. https://science.sciencemag.org/content/324/5923/81). The overarching goal is to understand better what in the merger history is most crucial in determining a galaxy’s present-day properties, an answer to which can be widely applicable to problems in galaxy formation.

Continue reading

Data is central to the NYC Department of Health’s mission to protect and promote the health of all New Yorkers. The agency’s many programs often require large scale record linkages that integrate data from individuals across multiple public health data systems and disease registries. We are implementing a Master Person Index (MPI) system in order to centralize, optimize and standardize matching methodology for administrative data across the Department of Health.

Continue reading

We are interested in investigating how deaths and hospitalizations resulting from opioid overdoses cluster across space and time in the US. This analysis will be conducted with the aid of two comprehensive databases: 1) detailed mortality data across the US; and 2) a stratified sample of all hospitalizations in the US, which can be subset to select for opioid overdoses. Analyses will be extended to drug type (prescription drugs, fentanyl etc.) and subject demographics (age, race, etc.). We have previously conducted similar cluster analysis for other health phenomena.

Continue reading

Through ArXivLab we aim to develop the next generation recommender systems for the scientific literature using statistical machine learning approaches. In collaboration with ArXiv we are currently developing a new scholarly literature browser which will be able to extract knowledge implicit in the mathematical and scientific literature, offer advanced mathematical search capabilities and provide personalized recommendations.

Continue reading

Effective representations and analyses of symbolic data, such as lexical data (words) and networks (graphs), have become of great interest in recent years, due both to advancements in data collection in Natural Language Processing (NLP), and the ubiquity of social networks. Such data often has no natural numerical representation, and is typically described in terms relational expressions or as pairwise similarities. It turns out that finding numerical representations of such data in “Hyperbolic” spaces—rather than into the more familiar Euclidean spaces—is a more effective way to preserve valuable relational information.

Continue reading