Search

Developing interpretable machine learning models has become an increasingly important issue. One way in which data scientists have been able to develop interpretable models has been to use dimension reduction techniques. In this paper, we examine several dimension reduction techniques including two recent approaches developed in the network psychometrics literature called exploratory graph analysis (EGA) and unique variable analysis (UVA). We compared EGA and UVA with two other dimension reduction techniques common in the machine learning literature (principal component analysis and independent component analysis) as well as no reduction to the variables real data. We show that EGA and UVA perform as well as the other reduction techniques or no reduction. Consistent with previous literature, we show that dimension reduction can decrease, increase, or provide the same accuracy as no reduction of variables. Our tentative results find that dimension reduction tends to lead to better performance when used for classification tasks.

The sudden arrest of motion due to confinement is commonly observed via the clogging transition in the flow of particles through a constriction. We present results of a simple experiment to elucidate a similar transition in the bidirectional flow of two species in which two species of macroscopic particles with different densities are confined in a tube and suspended in a fluid of intermediate density. Counterflowing grains serve as mobile obstacles and clogging occurs without arch formation due to confinement. We measure the clogging or jamming probability $J$ as a function of number of particles of each species $N$ in a fixed channel length for channel widths $D = $ 3$-$7$d$, where $d$ is the particle diameter. $J(N)$ exhibits a sigmoidal dependence and collapses on a single curve $J(N/D^3)$ indicating the transition occurs at a critical density. Data is well-fit by a probabilistic model motivated by prior constriction flows which assumes grains enter the clogging region with a fixed probability to produce a clogging state. A quasi-two-dimensional experiment provides insight into the interface shape and and we identify a Rayleigh-Taylor instability at large channel widths.

We calculate the chiral condensate of QCD at infinite coupling as a function of the number of fundamental fermion flavours using a lattice diagrammatic approach inspired by recent work of Tomboulis, and other work from the 80's. We outline the approach where the diagrams are formed by combining a truncated number of sub-diagram types in all possible ways. Our results show evidence of convergence and agreement with simulation results at small Nf. However, contrary to recent simulation results, we do not observe a transition at a critical value of Nf. We further present preliminary results for the chiral condensate of QCD with symmetric or adjoint representation fermions at infinite coupling as a function of Nf for Nc = 3. In general, there are sources of error in this approach associated with miscounting of overlapping diagrams, and over-counting of diagrams due to symmetries. These are further elaborated upon in a longer paper., Comment: presented at the 32nd International Symposium on Lattice Field Theory (Lattice 2014), 23-28 June 2014, New York, NY, USA

We develop a lattice diagrammatic technique for calculating the chiral condensate of QCD at infinite coupling inspired by recent work of Tomboulis and earlier work from the 80's. The technique involves calculating the contribution of gauge link diagrams formed from all possible combinations of a number of sub-diagram types. This is achieved by performing a resummation, using a truncated number of sub-diagram types. We show how to calculate the relevant sub-diagrams, including a new technique for evaluating group integrals with arbitrary number of gauge link elements, using Young Projectors. Including up to four different diagram types we calculate the chiral condensate as a function of Nf, and show that two real solutions result, which are non-zero for all integer Nf. We analyse these solutions and find signs of convergence of the expansion at small Nf. We discuss sources of error associated with this approach in detail and implement a technique to reduce over-counting of diagrams., Comment: 47 pages, including 2 appendices, 10 plots

We construct chiral perturbation theory for the gradient flow of the microscopic Dirac eigenvalues and compute the density of and correlations between the microscopic eigenvalues at zero and non-zero flow time. The results show that the repulsion of the microscopic Dirac eigenvalues from the dynamical quark mass decreases with increasing gradient flow time. Furthermore, the flow of the spectral resolvent is compared to the flow of the chiral condensate obtained from a fermionic gradient flow., Comment: 24 pages, 3 figures

We calculate an effective Polyakov line action of QCD at large Nc and large Nf from a combined lattice strong coupling and hopping expansion working to second order in both, where the order is defined by the number of windings in the Polyakov line. We compare with the action, truncated at the same order, of continuum QCD on S^1 x S^d at weak coupling from one loop perturbation theory, and find that a large Nc correspondence of equations of motion found in \cite{Hollowood:2012nr} at leading order, can be extended to the next order. Throughout the paper, we review the background necessary for computing higher order corrections to the lattice effective action, in order to make higher order comparisons more straightforward., Comment: 33 pages, 7 figures