Carnegie Mellon University


Guy Blelloch

Professor, Computer Science

Guy Blelloch works on parallel algorithms and programming languages. In the area of algorithms, he has worked on algorithms for a variety of problems including meshing, n-body codes, sorting, computational biology, graph problems, and compact data representations. In the area of programming languages, he has worked on developing new language structures for parallelism and techniques for efficiently executing parallel codes.

Personal web page:

Roy Briere

Professor, Physics

Roy Briere concentrates on precision measurements in weak flavor physics as a means for searching for new physics beyond the current Standard Model.  This method is complementary to direct searches for new particles at energy frontier machines.  The weak interactions of quarks are currently the only known source of matter-antimatter (CP) symmetry violation, but the preponderance of matter over antimatter in our universe tells us that there must be an additional source of CP violation that is as yet undiscovered.  He recently joined the BelleII Collaboration, which will continue the search for new CP-violating phenomena.

Rupert Croft

Professor, Physics

Rupert Croft’s main research interests are in computational cosmology, involving both simulations and the analysis of data from large surveys. This includes the physics of the intergalactic medium and its use as a probe of cosmology and of galaxy and quasar formation. He is participating in the SDSS surveys of galaxies and quasar absorption lines which are constraining dark energy, and is making the first "intensity mapping" measurements of structure using optical emission lines.  Croft also works on the re-ionization of the Universe, and high redshift galaxies, as well as new cosmological probes of modified gravity, such as gravitational redshifts and other relativistic effects which are just starting to be measured from galaxies and large-scale structure. He makes use of the McWilliams Center’s high performance computing facilities, including the Warp and Coma clusters to analyze SDSS data and perform cosmological hydrodynamic and radiative transfer simulations.

Tiziana Di Matteo

Professor, Physics

Tiziana Di Matteo is a theorist with expertise in both high energy astrophysics and cosmology.  Her interests focus on state-of-the-art cosmological simulations of galaxy formation with special emphasis on modeling the impact of black holes on structure formation in the Universe.  Her research makes extensive use of high-performance computing. Recently she has led an effort to run simulations of uniquely large volume and high resolution to study to the formation of the first large galaxies and quasars at the cosmic dawn of the Universe. This first population of galaxies and black holes will be investigated with the next generation telescopes (Euclid, JWST and WFIRST).  Large hydrodynamical cosmological simulations provide the direct link between the baryonic component and dark matter and are becoming useful in all stages of major observational projects in cosmology (Di Matteo is a member of the LSST Dark Energy Science Collaboration).

Scott Dodelson

Professor, Physics

Scott Dodelson proposes theories and analyzes data from cosmic surveys, focusing on questions such as: What is the dark matter? Is the dark energy vacuum energy? If so, why does it have such a peculiar value? Did inflation really happen? If so, is there any way to relate the fields that drove inflation to those we know about today? He serves as co-chair of the Science Committee for the Dark Energy Survey and is also involved in the South Pole Telescope and LSST Dark Energy Science Collaboration.

Christos Faloutsos

Professor, Computer Science

Christos Faloutsos works on data mining for large datasets. He has been using the idea of fractal dimension to characterize clouds of points in n-dimensional space, the 'hadoop'/mapReduce architecture to handle large datasets, and spectral methods to analyze large graphs.

Personal web page:

Peter Freeman

Project Scientist, Statistics

Peter Freeman's main research interest is the application of advanced statistical methods to astronomical data.  His recent work concentrates on advancing techniques for utilizing the information present in galaxy images (in collaboration with the HST CANDELS program) and for estimating photometric redshifts of galaxies (as a member of the LSST Dark Energy and LSST Informatics and Statistics Science Collaborations).

Personal web page:  

Christopher Genovese

Professor, Statistics

Christopher Genovese's research involves high-dimensional and nonparametric inferences with applications to complex scientific problems. A major focus of his research is developing statistical methods for problems in astrophysics and cosmology, including analysis of the Cosmic Microwave Background, inference for the dark energy equation of state, and source detection.

Personal web page:

Garth Gibson

Professor, Computer Science

Garth Gibson's research centers on large data processing and storage systems such as the world's largest parallel and distributed file systems for supercomputing and internet services. He leads a team that is exploring alternative parallel analysis computing models for large clusters (such as MapReduce) and applying these tools to astrophysics datasets that are terabytes to petabytes in size, and billions of particles in complexity.

Personal web page:

Fred Gilman

Professor, Physics

Fred Gilman’s research focused on understanding the nature of CP violation, which is a required ingredient in explaining the dominance of matter over antimatter in the universe. What is the nature of CP violation and of dark matter, dark energy, the field(s) responsible for inflation, and neutrino mass, constitute five unanswered questions about the universe which all involve particle physics that is not contained in the Standard Model and are addressable through cosmological observations.. Gilman is Director of the McWilliams Center for Cosmology, and a member of LSST Dark Energy Science Collaboration from the time of its creation.  He is the Chair of the AURA Management Council for the LSST (AMCL), the committee that oversees the construction and commissioning of the LSST Project, and is a member of the Board of Directors of the Association of Universities for Research in Astronomy (AURA).

Shirley Ho

Associate Professor, Physics

Shirley Ho is a cosmologist whose interest ranges from theory to observations, and whose research involves both simulations and analyses of large scale structure via novel techniques developed in Machine Learning and Statistics. Utilizing large scale structure and the cosmic microwave background, she seeks to understand the beginning of the Universe and its evolution, its dark components (dark energy and dark matter), and the light, elusive neutrinos. Her recent research focuses on the use of a standard ruler called Baryon Acoustic Oscillations via various large scale structure tracers such as the 3D clustering. In this way, she plays leading roles in large scale structure analyses in the SDSS-III, SDSS-IV, and Large Synoptic Sky Telescope collaborations (in particular, within the LSST Dark Energy Science Collaboration). In addition, she is a member of the future Dark Energy Spectroscopic Instrument (DESI) and Euclid surveys.

Tina Kahniashvili

Associate Research Professor, Physics

Tina Kahniashvili's research interests are in theoretical cosmology and astrophysics. Her research topics include testing physical processes in the early universe using cosmological and high energy astrophysical data (cosmic microwave background radiation, large scale structure, gamma ray bursts, blazars), modified theories of general relativity (in particular, massive gravity), the gravitational wave signal arising from inflation and cosmological phase transitions, dark energy-dark matter interacting cosmological models, primordial magnetic fields and their signatures, and magneto-hydrodynamic turbulence in astrophysical plasmas.

Leonard Kisslinger

Professor (emeritus), Physics

Leonard Kisslinger's astrophysics research is in the general area of astroparticle physics. Much of his research is on the electroweak and QCD phase transitions, which occurred when the temperature of the universe was about 100 Gev and 150 MeV, respectively. He has been investigating the magnetic fields which are produced for possible polarization correlations in the CMB radiation and as seeds for the galactic and extragalactic magnetic fields which have been observed. Recently, in collaboration with Professor Tina Kahniashvili, he has worked on the gravitational waves produced by these phase transitions. Also, in collaboration with experimentalists and theorists at LANL, he has been studying the detection of quark/gluon plasma in RHIC experiments and the very large velocities of pulsars that are produced in a supernova collapse.

Sergey Koposov

Assistant Professor, Physics

Sergey Koposov works on exploring the data from large astronomical surveys in order to answer various questions of galaxy formation and evolution. He is particularly interested in studying the Milky Way halo and different accretion events onto the Milky Way such as stellar streams. He also studies dwarf galaxies, star clusters and dark matter distribution in the local Universe. Sergey is involved in many large ongoing and future surveys such as Gaia, DESI, WEAVE and LSST.

Personal web page:

Ann Lee

Associate Professor, Statistics and Machine Learning

Ann Lee's interests are in developing statistical tools for complex scientific problems with high dimensions, heterogeneous structures and complex noise. Her current work (in collaboration with the CANDELS and LSST Dark Energy teams) involves comparing and estimating multivariate posterior distributions for galaxy images and photometric color data non-parametrically.

Personal web page:

Rachel Mandelbaum

Associate Professor, Physics

Rachel Mandelbaum's research interests are predominantly in the areas of observational cosmology and galaxy studies.  This work includes the use of weak gravitational lensing and other analysis techniques, with projects that range from development of improved data analysis methods, to actual application of such methods to existing data.  Currently, she is focusing on data from the SDSS (including SDSS-III and the ongoing SDSS-IV) and Hyper-SuprimeCam (HSC), and is working on upcoming surveys including LSST, Euclid, and WFIRST.

Manfred Paulini

Professor, Physics

Manfred Paulini studies questions connecting particle physics to issues relevant for cosmology. One such question is the predominance of matter over antimatter in the universe, which requires a breaking of the CP symmetry between matter and antimatter in particle physics. As a member of the CDF experiment at Fermilab, Paulini studied the violation of CP symmetry and matter-antimatter oscillations in neutral Bs mesons. Another question concerns the nature of dark matter that makes up about one quarter of the content of the universe. Paulini searches for the production of dark matter particles with the CMS experiment at the Large Hadron Collider at CERN. He analyzes the CMS data looking for supersymmetric particles with decay chains that involve photons in the final state.

Jeffrey Peterson

Professor, Physics

Jeff Peterson's group carries out cosmological observations using the 21 cm emission line of neutral hydrogen. The group pioneered the field of 21-cm Intensity Mapping using existing telescopes to make the first detection ofcosmic structure at redshifts near one. The team now contributes to the design of custom-built 21-cm telescopes in Canada, Mexico and China.  Currently, Peterson leads the RF design program for the HIRAX telescope in South Africa, an array of 1024 six-meter dishes slated for the South African Radio Astronomy Reserve. This telescope will map cosmic structure from redshift 0.8 to 2.5 allowing a sharp test of models of Dark Energy. These telescopes can also be used to study the mysterious, rare Fast Radio Bursts. The team recently reported the detection of the first convincingly extra-galactic radio burst.

Barnabas Poczos

Assistant Professor, Machine Learning

Barnabas Poczos is developing machine learning methods for advancing automated discovery and efficient data processing in applied sciences, including cosmology and astrophysics.  Currently, he is studying how to scale up learning algorithms to large, high-dimensional, complex datasets.  Poczos has developed non-parametric statistical methods to estimate differential entropy, divergences, mutual information, and other information theoretic quantities. These estimators can be used to generalize machine learning algorithms to be able to operate on complex objects such as functions, distributions, or sample sets.  The new learning methods have been applied to cosmology problems, including finding anomalous galaxy clusters, predicting the number of galaxies occupying a halo, estimating the cosmological parameters of our Universe, estimating the dynamical mass of galaxy clusters, and generating galaxy images conditioned on galaxy properties.

Personal web page:

Ira Rothstein

Professor, Physics

Ira Rothstein works on diverse topics in quantum field theory and general relativity.  He focuses on developing field theoretic tools for the purpose of increasing predictive power in complex non-linear field theories, such as QCD and gravity.  In the past, he has been particularly interested in developing effective field theory techniques in order to search for new physics in the Yukawa sector of the standard model.  Presently he has been concentrating on a recent formalism (NRGR) developed to study binary inspirals and using NRGR to calculate higher order post-Newtonian corrections to these systems for the purpose of building gravity wave templates for LIGO and LISA.  As with the rest of the particle physics community, he is anxiously awaiting the results from the LHC.  The advent of data collection will be a very exciting time, hopefully with the discovery of new physics that will explain the dark matter as well as the hierarchy problem.  Until then, he plans to work on finding methods to search for new physics in the model independent ways.  In particular, he is interested in understanding how to determine at early stages of LHC running if we have produced new particles but have not been able to extract them from the background.

James Russ

Professor, Physics

James Russ is a particle experimentalist whose work impacts the physics goals of the McWilliams Center through his involvement with the CMS experiment at the Large Hadron Collider and his work on neutrino astronomy as a probe of Active Galactic Nuclei (AGN) over different red-shift ranges. At CMS we have an opportunity to discover evidence of extra dimensions and indications of a brane-world, which would revolutionize cosmology. The neutrino astronomy project offers insight into nature's highest-energy particle sources and is one of the few ways to put experimental limits on models of AGN physics.

Chad Schafer

Associate Professor, Statistics

Chad Schafer's work focuses on addressing estimation problems in the sciences using novel, often computationally-intensive, statistical methods.  Projects in astronomy and cosmology have included the development of methods for constructing optimally precise confidence regions for cosmological parameters, for estimating bivariate luminosity functions, and for estimating properties of galaxies via low-dimensional representations of their emission spectra. Recent work has focused on the development of formal statistical estimation procedures that can take advantage of modern cosmological simulation models.

Personal web page:

Jeff Schneider

Associate Research Professor, Robotics Institute

Jeff Schneider is pursuing active learning for scientific discovery.  An active learning algorithm not only learns models from data, but also selects which experiments to run in order to collect the training data.  Recent work focused on algorithms that select cosmological parameters for CMBFast runs in order to find those consistent with the WMAP data (Bryan et al., ApJ, 2007).  This work has been extended to selecting across multiple model/data types (e.g., CMB, supernovae, large scale structure) with varying computational costs (Bryan and Schneider, ICML, 2008).  Other current efforts include approaching anomaly detection as an active learning problem and learning dynamic models from static data.

Personal web page:

Hy Trac

Associate Professor, Physics

Hy Trac is a theoretical and computational cosmologist whose scientific interests include cosmic evolution and structure formation. His work includes the development and application of numerical simulations to model and interpret the observable Universe. He is currently developing a novel mesh-free hydrodynamic code. In cosmology, he is especially interested in complex problems involving the gas, stars, galaxies, quasars, and clusters of galaxies that provide information about the underlying dark matter and dark energy. In astrophysics, he would particularly like to work on star and planet formation and the development of planetary atmospheres. He also collaborates with machine learning experts and statisticians to apply modern approaches to improve multi wavelength data analysis and numerical simulations. He is a member of the Atacama Cosmology Telescope (ACT) and Simons Observatory (SO) Collaborations.

Helmut Vogel

Professor, Physics

Helmut Vogel's main research interests are in high-energy particle physics experiments. For many years, he worked at LEP, the electron-positron colliding-beam accelerator which was the precursor to the LHC at CERN. Among its landmark results were a determination of the number of neutrino families in the universe and the most stringent lower limit to date on the mass of the Higgs boson. Presently, Vogel is a member of the CMS experiment at the LHC where he plans to study muons produced in the proton-proton collisions as probes of both "conventional" and "exotic" physics processes.

Matthew Walker

Assistant Professor, Physics

Matthew Walker studies the astrophysical properties of dark matter via optical imaging, spectroscopy, and dynamical modeling of the "dwarf" galaxies that surround the Milky Way. By measuring the small-scale clustering of dark matter, Walker aims to help figure out what the dark matter is. For this work he uses some of the world's largest optical telescopes, including the 6.5 meter Magellan telescopes at Las Campanas Observatory in Chile (where he is a member of the M2FS instrument team), the 6.5 meter MMT at Mt. Hopkins, Arizona, and the 8.2 meter Very Large Telescope at Cerro Paranal in Chile.

Larry Wasserman

Professor, Statistics and Machine Learning

Larry Wasserman's research interests include nonparametric interference, high-dimensional models, and the development of statistical methods for astrophysics problems such as: estimating the equation of state of dark energy; the analysis of the cosmic microwave background radiation; and filament finding.

Personal web page:

Postdoctoral Fellows

Duncan Campbell

McWilliams Postdoctoral Fellow, Physics

Duncan Campbell studies how galaxies relate to the large scale structure of dark matter in the Universe.  He is particularly interested in how the sites of galaxy formation, dark matter haloes, influence the properties of the galaxies they host.  For example, how do galaxies and dark matter haloes grow together, and what process(es) is responsible for shutting down star formation in some galaxies?  Understanding the connection between dark matter haloes and the galaxies they host is vital to interpret the large galaxy surveys that are used to constrain cosmological models.  In addition, he is interested in the computational challenges associated with modeling the next generation of galaxy surveys like LSST.  He received his PhD from Yale University in 2017 and is a member of the LSST Dark Energy Science Collaboration. 

Elena Giusarma

Postdoctoral Fellow, Physics

Elena Giusarma's major fields of research include: Neutrino and Axion Cosmology, Large Scale Structure, the Cosmic Microwave Background, Dark Energy and Dark Matter models, Inflationary models. In particular her work is focused on the study of dark matter properties and on the origin and evolution of the Universe using cosmological data.

Francois Lanusse

Postdoctoral Fellow, Physics

Francois Lanusse received his PhD at CEA Saclay under the supervision of Jean-Luc Starck.  His cosmological interests are in the areas of weak lensing and large-scale structure cosmology, using techniques from statistics (e.g., sparsity-based methods).

Danielle Leonard

McWilliams Postdoctoral Fellow, Physics

Danielle Leonard's research is focused on prospects for testing and constraining deviations from the standard cosmological model. She is particularly interested in alternative theories of gravity, but has worked on dynamical dark energy models and cosmology with spatial curvature. Her work has included forecasting parameter constraints, investigating theoretical uncertainties, and examining degeneracies, with a focus on the cosmological observable of weak gravitational lensing. She is also interested in weak lensing more generally, including the mitigation of systematic effects such as intrinsic alignments. She is a member of the LSST Dark Energy Science Collaboration.

Personal web page:

Markus Michael Rau

Postdoctoral Fellow in Astrophysics and Machine Learning, Physic

Markus Michael Rau is interested in using techniques from Machine Learning and Statistics to answer fundamental questions about the formation and evolution of structure in the Universe. Large area, data-intensive, photometric surveys like LSST, KiDS or DES, will strongly increase the statistical precision of cosmological measurements in the following years. These projects will enable us to investigate the nature of dark energy and the formation of structure to an unprecedented accuracy. In the advent of these exciting prospects for cosmological research, Markus innovates novel methodologies to analyze these large observational datasets. During his Ph.D., he developed techniques to accurately measure the redshift of galaxies based on their photometry, which is a vital prerequisite for an accurate cosmological analysis. In this context, he is particularly interested in techniques that incorporate sources of systematic error, like photometric redshift uncertainty, into cosmological parameter constraints. The rapid development of Computer Science and particularly Machine Learning offers exciting opportunities for Astrophysics and Cosmology. Markus is particularly interested in finding new ways to use Machine Learning in concert with scientific intuition for the benefit of cosmological research. Markus Rau received his Ph.D. from the Ludwig Maximilians University in Munich in 2017. He is a member of the Dark Energy Survey collaboration.

Simon Samuroff

Postdoctoral Fellow, Physics

Simon Samuroff studies the large scale properties of the Universe using weak lensing and other late-time probes. His interests broadly fall under the label of data-led cosmology. Specifically he is interested in understanding the complementary constraining power of different observational datasets, and the extent to which they can be used to mitigate lensing systematics. His recent work focuses on using real data from the Dark Energy Survey and hydrodynamic simulations to understand the nature, scale and dependence on galaxy properties of intrinsic galaxy shape correlations. He also works on the delicate problem of galaxy shape measurement and its biases, using image simulations to explore the effects of blending. He is particularly interested in the development of new methods for deblending, how different shear estimation techniques respond to the presence of blending and one should interpret the results. He received his PhD in 2017 from the University of Manchester. He is a member of the Dark Energy Survey and the LSST Dark Energy Science Collaboration.

Mei-Yu Wang

McWilliams Postdoctoral Fellow, Physics

Mei-Yu Wang is interested in utilizing various astrophysical probes to study the nature of dark matter. She is particular interested in Milky Way substructure properties and evolution histories, but also works on astrophysical uncertainties on direct and indirect dark matter search, and high-redshift structure formation such as Lyman-alpha forest. Much of her work involves comparing small-scale dark matter clustering predictions from cosmological simulations and dynamical modeling with large astronomical data set.​

Visiting Researchers

Ross O'Connell

Visiting Research Scholar, Physics

Ross O’Connell uses the large-scale-structure of the universe to study fundamental physics.  His research interests include dark energy, primordial non-gaussianity, and large scale tests of general relativity.  He works primarily with the BOSS and eBOSS surveys, with current projects involving both the Lyman-alpha forest and luminous red galaxy portions of those surveys.