Date of Degree


Document Type


Degree Name





Adrian Dumitru

Committee Members

Stefan Bathe

Jamal Jalilian-Marian

Andrea Ferroglia

Brian C. Tiburzi

Bjoern Schenke

Subject Categories

Elementary Particles and Fields and String Theory | Nuclear | Quantum Physics


Reweighting, Color Glass Condensate, Small-x, QCD, Gluons, Double Inclusive Gluon Production


In this dissertation, I will argue that we can study functional fluctuations in unintegrated gluon distributions, in the MV model as well as JIMWLK, using reweighting techniques, which will allow me to calculate QCD observables with "biased ensembles". This technique will enable me to study rare functional configurations of the gluon distributions, that might have been selected for in, for example, the centrality criteria used by the ATLAS and ALICE collaborations. After a review of these techniques, as well as a review of QCD physics at high energy in general, I will use biased ensembles to compute observables in two examples.

First, I’ll look at the nuclear modification factor in pA/pp collisions. Recent experimental data suggests that the nuclear modification factor should increase with centrality. I show (based on a paper written with collaborators Adrian Dumitru and Vladimir Skokov) that a functional bias on gluon distributions does, indeed, bare this out.

My second application will use biased ensembles to compute cross sections and azimuthal harmonics in double inclusive gluon production via “Glasma graphs”. I use three types of biases (an increase in gluons multiplicity in some momentum range by a constant factor; by a factor that is momentum dependent; and by a factor that is anisotropic) to show that biased ensembles introduce interesting and diverse changes in the angular anisotropies. Along the way, I'll stress that one should always remember how likely each bias that is being studied is, in the original ensemble (which, after all, is what these techniques are studying).

My hope for this work is that, not only will it provide an additional way of interpreting experimental results using saturation physics, but that it will help the field in some way, as it looks onward to, for example, the EIC expansion at Brookhaven.