Prof. Dvorkin is a theoretical cosmologist. She has made significant contributions to the study of dark matter (DM), light relics, and the physics of the early universe, focusing on a broad range of observational probes, such as the Cosmic Microwave Background (CMB), the large-scale structure of the universe, and strong gravitational lensing.

She has pushed the frontiers of sub-GeV dark matter-baryon scattering using cosmic microwave background (CMB) and large-scale structure (LSS) data, which provide complementary constraints to current direct detection experiments. This scenario became one of the main drivers for the DM science case of the next-generation CMB-S4 experiment.

With her research group, they have developed a general formalism aimed at probing DM at small cosmological scales using gravitational lensing, by means of a statistical measurement of dark matter substructure. They have demonstrated with high-resolution simulations the ability of such a measurement to discern between different dark matter scenarios. They have also quantified the contribution from line-of-sight halos and subhalos and found the former to be dominant for many of the strong lensing systems studied in the literature. Inspired by this, they reanalyzed a substructure in the JVAS B1938+666 lens (previously claimed to be a subhalo) and showed, with decisive evidence, that it is instead a halo along the line-of-sight. This constitutes the first dark perturber shown to be a line-of-sight halo with a gravitational lensing method. Her group showed, for the first time in the literature, that machine learning techniques can accelerate direct detection of dark matter perturbers in lensing systems by many orders of magnitude.

In collaboration with particle physicists, Dvorkin identified a new production channel for DM produced through the freeze-in mechanism: the decay of photons that acquire an in-medium plasma mass. This turned out to be a dominant channel for sub-MeV DM production, and including this channel leads to a significant reduction in the predicted signal strength for current and upcoming direct detection searches. Dvorkin computed for the first time in the literature the full non-thermal velocity distribution of the DM in this type of scenarios with DM-baryon interactions and ruled out masses below 20 keV, using CMB measurements by the Planck satellite.

In recent work with her group, Dvorkin proposed and conducted a systematic search for Light (but Massive) Relics (LiMRs), using the latest CMB, weak lensing and galaxy data sets. Light relics are a generic prediction of a broad range of beyond-SM (BSM) scenarios and could compose (part of) the dark matter. They are commonly assumed to be massless, but this does not need to be true. Examples of LiMRs include neutrinos, axions, gravitinos, dark photons, etc. Her group set the tightest and most comprehensive constraints to date on eV-scale bosonic and fermionic relics. The search for LiMRs complements current efforts in the search for (massless) relics with the CMB, and opens a new route for finding BSM physics. This body of work was used by the Dark Energy Survey collaboration to look for BSM physics in their most recent cosmological analysis and it prompted an invitation to lead the Snowmass white paper on light relics.

Recently, her group performed the first field-level analysis of a galaxy data set. They used the wavelet scattering transform (WST), originally proposed by mathematicians 10 years ago, to analyze BOSS data, and found improvements in the constraints on cosmological parameters over the ones coming from the standard power spectrum analysis. The WST presents several advantages over traditional estimators (in that it is more efficient) and over “black box” machine learning methods (in that it is interpretable). Further work is ongoing.

In the early universe arena, she has pioneered a model-independent formalism for probing the shape of the inflationary potential, known as "Generalized Slow Roll". This formalism has been widely used in the literature for primordial features studies and it has been applied to the data by several members of the community, including the Planck collaboration. The CMB-S4 collaboration plans to use this approach as its main way of reconstructing the shape of the inflationary potential. She has also constructed new theoretical templates for higher-order correlation functions of the initial curvature perturbations that could shed light on the physical properties of particles with non-zero spin during inflation as well as possible phase transitions during the early universe. She developed statistical tools to look for these correlation functions in the CMB and the LSS data measured by current and future surveys. 

In 2014, she joined the joint analysis between BICEP2, the Keck array, and Planck collaboration. She worked on the likelihood analysis of a multi-component model that included Galactic foregrounds and a possible contribution from inflationary gravity waves. The code that she wrote was made publicly available, and it has been extensively used by the community. No statistically significant evidence for primordial gravitational waves and strong evidence for galactic dust were reported in this work.

While studying the observational imprints of inflation on the CMB, Dvorkin became interested in the period of reionization. She developed a new statistical technique for extracting the inhomogeneous reionization signal from measurements of the CMB polarization. This technique has been tested in actual data and implemented by several members of the community. She further showed that existing calculations of the B-mode polarization power spectrum from reionization were incomplete by finding an additional signal of the same order of magnitude as the one that had been previously calculated. These B-modes have been sought for and seen in simulations by several groups.

Prof. Dvorkin is the Harvard Representative at the NSF-funded Institute for Artificial Intelligence and Fundamental Interactions (IAIFI)’s Board. This is a joint effort together with colleagues at Harvard, MIT, Tufts, and Northeastern.

She was the co-leader of the Inflation analysis group for the next-generation CMB-S4 experiment. Prior to this, she was the leader of the Dark Matter analysis group. She is also a full member of the Vera Rubin Observatory’s LSST Dark Energy Science Collaboration (DESC).

In 2022, Dvorkin was voted “favorite professor” by the Harvard senior Class of 2023. She has been awarded the 2019 DOE Early Career award and has been named the "2018 Scientist of the year" by the Harvard Foundation for "Salient Contributions to Physics, Cosmology and STEM Education" (with support from Harvard students). She has also been awarded a Radcliffe Institute Fellowship for 2018-2019 and a Shutzer Professorship at the Radcliffe Institute for the period 2015-2019. In 2018 she was awarded a Star Family Challenge prize for Promising Scientific Research, which supports high-risk, high-impact scientific research at Harvard. In 2012, she was given the "Martin and Beate Block Award", awarded to the best young physicist by the Aspen Center for Physics.

Professor Dvorkin, born and raised in Buenos Aires, Argentina, received her Diploma in Physics from the University of Buenos Aires with honors. She earned her Ph.D. in the Department of Physics at the University of Chicago in 2011, where she won the "Sydney Bloomenthal Fellowship for "outstanding performance in research". She has conducted postdoctoral research at the School of Natural Sciences at the Institute for Advanced Study in Princeton (from 2011 to 2014) and at the Institute for Theory and Computation at the Center for Astrophysics at Harvard University (from 2014 to 2015), where she was both a Hubble Fellow (awarded by NASA) and an ITC fellow, before becoming a professor at the Physics Department at Harvard University.

___________________________
Faculty Assistant: Morgan Holly