ENTROPY IN QUANTUM MECHANICS:

RECENT ADVANCES
JUNE 25TH, 26TH 2013
Universit de Cergy-Pontoise site St Martin, espace Colloques

-----

TUESDAY 25TH
9:30 - 10:00 Coffee
10:00 - 10:10 Opening
10:10 - 11:00 Denes PETZ (Budapest): Some inequalities in probabilities and matrices
11:10 - 12:00 Jan Philip SOLOVEJ (Copenhagen): Solution to a conjecture on the classical entropy of
quantum states
12:10 - 14:00 Lunch
14:00 - 14:50 Matthias CHRISTANDL (Zurich): Recoupling Coefficients and Quantum Entropies
15:00 - 15:50 Fernando BRANDAO (Belo Horizonte): Exponential Decay of Correlations Implies Area Law
16:00 - 16:30 Coffee Break
16:30 - 17:20 Eric CARLEN (Rutgers): Non-commutative Optimal mass transportation and gradient flow
-----

WEDNESDAY 26TH
9:30 - 10:00 Coffee
10:00 - 10:50 Renato RENNER (Zurich): A proof of the data processing inequality based on smooth entropy
11:00 - 11:50 Frank HANSEN (Tohoku): Convex trace functions with applications in quantum physics
12:00 - 14:00 Lunch
14:00 - 14:50 Frank VERSTRAETE (Vienna): Entanglement rates and area laws
15:00 - 15:30 Elliott H. LIEB (Princeton): Monotonicity of a relative Renyi entropy
15:30 - 16:00 Coffee

ABSTRACTS
Denes PETZ: Some inequalities in probabilities and matrices
There are contacts about probability theory, von Neumann entropy, relative entropy, Tsallis entropy and variance of matrices. The lecture
includes rather recent results including my papers and I suggest a few possible new results.
Jan Philip SOLOVEJ: Solution to a conjecture on the classical entropy of quantum states
It is a celebrated fact that quantum states have an inherent uncertainty often expressed through the Heisenberg uncertainty relation.
It is therefore natural, but not entirely straightforward, to introduce the classical entropy of a quantum state. Using the coherent state
transform Wehrl (1979) suggested such a definition of classical entropy for a quantum state on the standard quantization of Euclidean
space. He formulated the natural conjecture that this classical entropy is minimized by states that also minimize the Heisenberg uncertainty
inequality, i.e., Gaussian coherent states. Using the sharp form of Young's and Hausdorff-Young's inequalities Lieb (1978) proved this
conjecture. Lieb moreover conjectured that the same should be correct for the quantization of the sphere. This corresponds to all the
irreducible SU(2) representations, where the Euclidean Glauber coherent states are replaced by SU(2) Bloch coherent states (maximal
weight vectors). This generalized Wehrl conjecture has been open for almost 35 years. I will present a short proof, which is joint work with
Elliott Lieb. The conjecture may be generalized to other compact semisimple Lie groups, but here it is still open.
Matthias CHRISTANDL: Recoupling Coefficients and Quantum Entropies
In this work, we show that the asymptotic limit of the recoupling coefficients of the symmetric group is characterized by the existence of
quantum states of three particles with given eigenvalues for their reduced density matrices. This parallels Wigner's observation that the
semiclassical behavior of the 6j-symbols for SU(2)---fundamental to the quantum theory of angular momentum---is governed by the
existence of Euclidean tetrahedra. We explain how to deduce solely from symmetry properties of the recoupling coefficients the strong
subadditivity of the von Neumann entropy, first proved by Lieb and Ruskai, and discuss possible generalizations of our result. This is joint
work with Burak Sahinolgu and Michael Walter, available at http://arxiv.org/abs/1210.0463 .
Fernando BRANDAO: Exponential Decay of Correlations Implies Area Law
Quantum states of many particles are fundamental to our understanding of many-body physics. Yet they are extremely daunting objects,
requiring in the worst case an exponential number of parameters in the number of subsystems to be even approximately described. How
then can multi-particle states be useful for giving predictions to physical observables? The intuitive explanation is that physically relevant
quantum states, defined as the ones appearing in nature, are usually much simpler than generic quantum states.
In this talk I will discuss a very recent theorem that gives further justification to this intuition.
The theorem states that exponential decay of correlations, a physically motivated restriction on the set of multi-particle quantum states,
implies an area law for the entanglement entropy of systems defined on a line, and thus also an efficient classical description for such
systems. The result can be seen as a rigorous justification to the intuition that states with exponential decay of correlations, usually
associated with non-critical phases of matter, are simple to describe.
I will outline the main steps in the proof, that relies on several previous tools from quantum information theory and that can also be seen as
providing a limitation to the phenomenon of data hiding in quantum states. Based on joint work with Michal Horodecki.
Eric CARLEN: Non-commutative Optimal mass transportation and gradient flow
We introduce a metric on the space of density matrices for an N- fermion system that is the analog of the 2-Wasserstein metric. In particular,
we show that the fermionic Fokker-Planck evolution is gradient flow for the von Neumann entropy in this metric, and we prove a Talagrand
inequality for this metric. Several open problems are discussed. This is joint work with Jan Maas.
Renato RENNER: A proof of the data processing inequality based on smooth entropy
Smooth entropy has been introduced recently as an operationally motivated measure of uncertainty. In this talk, I will show --- by an
elementary argument --- that smooth entropy satisfies a data processing inequality. Since smooth entropy can be seen as a generalisation
of von Neumann entropy (and therefore reduces to the latter for appropriately chosen parameters) this gives a novel proof for the data
processing inequality for the von Neumann entropy, and therefore for its strong subadditivity.
Frank HANSEN: Convex trace functions with applications in quantum physics
We introduce new classes of convex trace functions with applications in quantum physics. In particular, we obtain convexity statements for t
he residual entropy, the entropy gain over a single quantum channel, and the entropy gain over multiple channels. In addition, we propose n
ew methods based on the theory of operator monotone and operator convex functions to study Minkowski type trace inequalities.
Frank VERSTRAETE: Entanglement rates and area laws
Optimal bounds will be discussed on the rate at which the entanglement entropy of a subsystem can change by Hamiltonian evolution. As
an application, it will be shown that the area law for the entanglement entropy is well defined and stable within a gapped phase.
Elliott H. LIEB: Monotonicity of a relative Renyi entropy
We show that a recent definition of relative Renyi entropy is monotone under completely positive, trace preserving maps. This proves a
recent conjecture of Mller--Lennert et al. The ingredients in our proof are Lieb's concavity theorem, Ando's convexity theorem and
theorems of Carlen--Lieb. Joint work with Rupert Frank.