Highlights of recent research:
Dark Matter Dynamics and Statistics
We have been pioneering non-linear transformations that reveal a deep, hitherto unknown connection of dynamics and statistics in the non-linear regime. We transferred advanced techniques from mathematical statistics to show that the transformations we found are indeed optimal from Fisher information point of view, they Gaussianize the distributions, “undo” non-linear evolution, and open up information content that is not even accessible when measuring infinite (!) number of moments (the mainstream thinking is measuring a few more moments). Along with the techniques we developed reliable forecasting tools, and showed that we are able to increase the pay off of WFIRST, EUCLID, HSC, etc. in terms of Figure of Merit by about factor of two, and in simulations by factors of up to 40. In addition, we managed understand the role of statistics and dynamics in the NFW profiles: indeed, if a dynamical origin of a potential floor is found in the center of dark matter halos, our theory will provide full understanding. Most recently, we invented (yet another) series of new statistics, split and sliced power spectra that work as a microscope unraveling the density dependence of Dark Matter growth. Our first result will show how the hitherto puzzling shift of the BAO scale is caused by the largest peaks in the density field. New results on alternative gravity, redshift distortion are imminent.
CMB Anomalies and Cold Spot
We have been investigating the imprint of local large scale structure on the CMB and in particular, whether they might produce the CMB anomalies, puzzling phenomena in tension with the standard LCDM theory. The largest statistically significant structure identified in the Universe, a previously hypothesized super-void, has been found by my group. This explains to the most significant of anomalies on the CMB, the Cold Spot.
Analysis of large astronomical data sets
Along the the statistics developed in projects above, we invented a slew of new algorithms that enable the efficient realization of these methods, as well as new compression paradigms for saving huge N-body simulations.