New Techniques in Deep Representation Learning Galen Andrew Chair of the Supervisory Committee: Associate Professor Emanuel Todorov CSE, joint with AMATH The choice of feature representation can have a large impact on the success of a machine learning algorithm at solving a given problem. Although human engineers employing taskspecific domain knowledge still play a key role in feature engineeri...

We report the discovery of a quasi-periodic oscillation (QPO) at 642 mHz in an XMM–Newton observation of the ultraluminous X-ray source (ULX) IC 342 X-1. The QPO has a centroid at νQPO = 642 ± 20 mHz, a coherence factor Q = 11.6 and an amplitude (rms) of 4.1 per cent with significance of 3.6σ . The energy dependence study shows that the QPO is stronger in the energy range 0.3–5.0 keV. A subsequ...

Energy-momentum is an important conserved quantity whose definition has been a focus of many investigations in general relativity. Unfortunately, there is still no generally accepted definition of energy and momentum in general relativity. Attempts aimed at finding a quantity for describing distribution of energy-momentum due to matter, non-gravitational and gravitational fields only resulted i...

We compare the performance of several robust large-scale minimization algorithms applied for the minimization of the cost functional in the solution of ill-posed inverse problems related to parameter estimation applied to the parabolized Navier-Stokes equations. The methods compared consist of the conjugate gradient method (CG), Quasi-Newton (BFGS), the limited memory Quasi-Newton (L-BFGS) [1],...

The paper presents concrete realizations of quasi-Newton methods for solving several standard problems including complementarity problems, special variational inequality problems, and the Karush–Kuhn–Tucker (KKT) system of nonlinear programming. A new approximation idea is introduced in this paper. The Q-superlinear convergence of the Newton method and the quasiNewton method are established und...