Condensed Matter

Strongly interacting quantum systems driven out of equilibrium represent a fascinating field where several questions of fundamental importance remains to be addressed [1].

These range from the dynamics of high-dimensional interacting models to the thermalization properties of quantum gases in continuous space.

In this Seminar I will review our recent contributions to some of the dynamical quantum problems which have been traditionally inaccessible to accurate many-body techniques.

I will first focus on the main methodological developments we devised in the past years.

In particular, I will describe the time-dependent Variational Monte Carlo method [2,3] and two notable classes of variational quantum states : the time-dependent Jastrow-Feenberg expansion, and the most recently introduced Neural-network Quantum States [4]. These states can achieve high (and controllable) accuracy both in one and higher dimensions.

Then, I will discuss specific applications to the problem of information spreading in both short- and long-ranged interacting quantum systems [3,5]. Finally, I will also discuss recent applications to thermalization properties of Lieb-Liniger quantum gases [6].

Tensor networks have been very successful for approximating quantum states that would otherwise require exponentially many parameters.

I will discuss how a similar compression can be achieved in models used to machine learn data, such as sets of images, by representing the fitting parameters as a tensor network. The resulting model achieves state-of-the-art performance on standard classification tasks. I will discuss implications for machine learning research, exploring which insights from physics could be imported into this field.

I will discuss the stability and breakdown of the topological classification of gapped ground states of non-interacting fermions, the tenfold way, in the presence of quartic fermion-fermion interactions.  In our approach [1], the effects of interactions on the boundary gapless modes are encoded in terms of boundary dynamical masses. Breakdown of the non-interacting topological classification occurs when the quantum nonlinear sigma models for the boundary dynamical masses favor quantum disordered phases.  The non-interacting topological classification $Z$ in odd spatial dimensions is unstable and reduces to $Z_N$ that can be identified explicitly for any dimension and any defining symmetries.
[1] T. Morimoto, A. Furusaki, and C. Mudry, Phys. Rev. B 91, 235111 (2015).

I will review recent progress on theory of many-body localization, mostly focusing on properties of the many-body localized phase itself.

I will discuss explicit construction of effective Hamiltonians governing the dynamics of conserved quantities.  The analysis reveals several inequivalent length scales in the system, some of which do not appear to diverge on the approach to the thermalized phase.

Experimental protocols to measure these length scales will also be discussed.

Numerical results suggest that the quantum Hall effect at {\nu} = 5/2 is described by the Pfaffian or anti-Pfaffian state in the absence of disorder and Landau level mixing. Those states are incompatible with the observed transport properties of GaAs heterostructures, where disorder and Landau level mixing are strong. We show that the recent proposal of a PH-Pfaffian topological order by Son is consistent with all experiments. The absence of the particle-hole symmetry at {\nu} =

5/2 is not an obstacle to the existence of the PH-Pfaffian order since the order is robust to symmetry breaking.