[論文レビュー] Concentration of quantum states from quantum functional and transportation cost inequalities

Cambyse Rouzé, Nilanjana Datta|arXiv (Cornell University)|Apr 7, 2017

Spectral Theory in Mathematical Physics参考文献 55被引用数 52

ひとこと要約

本論文は、quantum Wasserstein distances を用いて量子輸送コスト不等式 TC1 および TC2 を定義し、それらを MLSI および Poincaré 不等式と関連付け、恒等量子状態の濃度結果を導出する。 depolarizing semigroups および有限ブロック長量子パラメータ推定への適用も含む。

ABSTRACT

Quantum functional inequalities (e.g. the logarithmic Sobolev- and Poincar\\'e inequalities) have found widespread application in the study of the behavior of primitive quantum Markov semigroups. The classical counterparts of these inequalities are related to each other via a so-called transportation cost inequality of order 2 (TC2). The latter inequality relies on the notion of a metric on the set of probability distributions called the Wasserstein distance of order 2. (TC2) in turn implies a transportation cost inequality of order 1 (TC1). In this paper, we introduce quantum generalizations of the inequalities (TC1) and (TC2), making use of appropriate quantum versions of the Wasserstein distances, one recently defined by Carlen and Maas and the other defined by us. We establish that these inequalities are related to each other, and to the quantum modified logarithmic Sobolev- and Poincar\\'e inequalities, as in the classical case. We also show that these inequalities imply certain concentration-type results for the invariant state of the underlying semigroup. We consider the example of the depolarizing semigroup to derive concentration inequalities for any finite dimensional full-rank quantum state. These inequalities are then applied to derive upper bounds on the error probabilities occurring in the setting of finite blocklength quantum parameter estimation.

研究の動機と目的

• Extend classical transportation-cost and functional inequalities to the quantum setting using quantum Wasserstein distances.
• Investigate the relationships between quantum TC1, TC2, MLSI, and Poincaré inequalities and their implications for concentration.
• Obtain concentration bounds (Gaussian and exponential) for the invariant state of quantum Markov semigroups.
• Apply the concentration results to quantum state parameter estimation in finite blocklength regimes.

提案手法

• Introduce quantum Wasserstein distances of order 1 and 2 using appropriate operator tools and modular theory.
• Define Dirichlet forms and quantum gradient/divergence to formulate quantum MLSI and Poincaré inequalities.
• Prove a chain of implications: TC2 implies TC1; MLSI implies TC2; TC2 implies Poincaré; and derive corresponding concentration results.
• Establish quantum de Bruijn identity linking entropy production to Dirichlet forms.
• Specialize to primitive quantum Markov semigroups with full-rank invariant state and provide structural form (GKLS) of generators.
• Explore the depolarizing semigroup as a concrete example to illustrate concentration for full-rank states.

実験結果

リサーチクエスチョン

• RQ1 How can TC1 and TC2 be defined in the quantum setting using quantum Wasserstein distances?
• RQ2 How do quantum TC1/TC2 relate to quantum MLSI and quantum Poincaré inequalities?
• RQ3 Do quantum transportation-cost inequalities imply concentration phenomena for the invariant state of a quantum Markov semigroup?
• RQ4 What are concrete quantum examples (e.g., depolarizing semigroup) that illustrate these quantum concentration results?
• RQ5 How can these results be leveraged to bound finite-blocklength quantum parameter estimation errors?

主な発見

• Quantum analogues of TC1 and TC2 are defined and related to MLSI and Poincaré inequalities in the quantum setting.
• A chain of implications is established: TC2 implies TC1; MLSI implies TC2; TC2 implies Poincaré (PI).
• The framework yields Gaussian and exponential concentration bounds for the invariant state of quantum Markov semigroups.
• Concentration results are illustrated via the depolarizing semigroup, yielding bounds for finite-dimensional full-rank quantum states.
• Applications to finite blocklength quantum parameter estimation are developed, providing upper bounds on error probabilities.