Article
Statistics & Probability
Matey Neykov, Sivaraman Balakrishnan, Larry Wasserman
Summary: The study focuses on the problem of conditional independence testing of X and Y given Z, considering smoothness assumptions on conditional distributions and testing difficulty. Lower and upper bounds were derived on the critical radius of separation between null and alternative hypotheses in the total variation metric.
ANNALS OF STATISTICS
(2021)
Article
Physics, Multidisciplinary
Andrew Rolph
Summary: In this article, we study the entanglement structure and dynamics in CFTs and black holes, employing a local entanglement measure known as the entanglement contour. We calculate the entanglement contour for different systems, including 1+1-dimensional condensed matter systems and simplified models of black hole evaporation. Our findings reveal universal results for the entanglement contours in low energy non-equilibrium states of 2D CFTs and illustrate the presence of an island phase transition in the entanglement contour of a non-gravitational bath coupled to an extremal AdS(2) black hole.
Article
Quantum Science & Technology
Maurice Weber, Nana Liu, Bo Li, Ce Zhang, Zhikuan Zhao
Summary: Quantum machine learning models have the potential to be faster and more accurate than classical models, but they are also vulnerable to input perturbations. A fundamental link between binary quantum hypothesis testing and provably robust quantum classification has been formalized, leading to a tight robustness condition that puts constraints on the amount of noise a classifier can tolerate. This robustness condition against worst-case noise scenarios extends to known noise sources, providing a framework to study the reliability of quantum classification protocols beyond adversarial attacks.
NPJ QUANTUM INFORMATION
(2021)
Article
Physics, Multidisciplinary
Han Xu, Benran Wang, Haidong Yuan, Xin Wang
Summary: Quantum hypothesis testing is crucial in quantum technologies, and improving the robustness of quantum control is essential for accurate testing.
NEW JOURNAL OF PHYSICS
(2023)
Article
Physics, Multidisciplinary
Marta Maria Marchese, Alessio Belenchia, Stefano Pirandola, Mauro Paternostro
Summary: Quantum hypothesis testing applied to optomechanical systems reveals advantages in discriminating competing hypothesis, particularly in utilizing input squeezed optical noise and feasible measurement schemes on output cavity modes. This provides an advantage over classical schemes and offers possibilities for fundamental physics searches, especially in discriminating models of spontaneous collapse of the wavefunction.
NEW JOURNAL OF PHYSICS
(2021)
Article
Physics, Multidisciplinary
Shijie Cui, Junqing Li, Li Huang
Summary: In this paper, we propose a new bipartite entanglement monotone based on the minimum relative entropy of any bipartite quantum entanglement state. We also demonstrate that this entanglement monotone satisfies some basic properties as an entanglement measure.
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2023)
Article
Computer Science, Theory & Methods
Yongjune Kim, Cyril Guyot, Young-Sik Kim
Summary: This paper introduces two min-entropy estimators based on Coron's test and Kim's test to improve computational complexity and estimation accuracy. By analyzing the bias-variance tradeoff, we observe that an order of two is appropriate, focusing on min-entropy estimation from collision entropy.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2021)
Article
Computer Science, Information Systems
Lianyi Yu, Qiangjiang Wang, Yan Wo, Guoqiang Han
Summary: Biometric hashing is widely used in privacy-preserving biometric recognition systems due to its irreversibility, low computational cost, and high storage efficiency. However, the security of biometric hashing has been challenged by relation-based attacks. To address this issue, researchers propose a Secure Biometric Hashing scheme against Relation-Based Attacks (SBH-RA), which minimizes the leakage of distance relation on the original biometric by maximizing the conditional min-entropy of the signs of inter-class distance differences.
COMPUTERS & SECURITY
(2022)
Article
Statistics & Probability
Nikolaj Thams, Sorawit Saengkyongam, Niklas Pfister, Jonas Peters
Summary: This study introduces statistical testing under distributional shifts. It addresses the problem of testing a hypothesis about a target distribution when data is observed from a different distribution. The proposed method resamples from the observed data and applies an existing test in the target domain, inheriting its asymptotic level and power.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
(2023)
Article
Public, Environmental & Occupational Health
Hye Kyung Kim, Sungeun Chung, Youllee Kim, Seoin Lee
Summary: This study examined the relationship between perceived risk and protection behavior, and found support for both the behavioral motivation hypothesis and the risk reappraisal hypothesis. Higher risk perception motivates protection behaviors, while protection behaviors reduce perceived risk.
SOCIAL SCIENCE & MEDICINE
(2022)
Article
Physics, Multidisciplinary
Charlotte K. Duda, Kristina A. Meier, Raymond T. Newell
Summary: We introduce a quantum random number generator (QRNG) that operates in a PCI express form factor-compatible plug-and-play design. The QRNG utilizes a thermal light source with photon bunching following the Bose-Einstein statistics. We demonstrate that 98.7% of the unprocessed random bit stream min-entropy can be attributed to the quantum signal. The classical component is then removed using a non-reuse shift-XOR protocol, and the final random numbers pass various statistical randomness test suites.
Article
Computer Science, Information Systems
Michael G. Jabbour, Nilanjana Datta
Summary: We prove a tight uniform continuity bound for Arimoto's version of the conditional alpha-Renyi entropy for the range alpha is an element of [0, 1). The conditional alpha-Renyi entropy is a natural and widely used concept in information theory, and it has applications in various tasks such as guessing and decoding. Our result also reveals the relationship between the conditional alpha-Renyi entropy and the conditional Shannon entropy.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2022)
Article
Physics, Multidisciplinary
Uwe Hohm, Christoph Schiller
Summary: This paper reviews the experimental and theoretical results on the entropy limits for macroscopic and single-particle systems. All experiments confirm the minimum system entropy of S > kln2. The paper clarifies the cases where it is possible to discuss the minimum system entropy kln2 and the cases where the quantum of entropy can be discussed. It resolves conceptual tensions with the third law of thermodynamics, the additivity of entropy, statistical calculations, and entropy production. The paper also surveys black hole entropy. Claims for smaller system entropy values are shown to contradict the requirement of observability, and it is argued for the first time that this also implies the minimum system entropy kln2. The uncertainty relations involving the Boltzmann constant and the derivation of thermodynamics from the existence of minimum system entropy enable the discussion of a general principle valid across nature.
Article
Physics, Multidisciplinary
Nicolas P. Bauer, Jan Carl Budich, Alessio Calzona, Bjoern Trauzettel
Summary: We propose a novel spatially inhomogeneous setup to study the effects of quench-induced fractionalized excitations in entanglement dynamics. By coupling a region undergoing quantum quench to a static probe region, we can monitor the time-dependent entanglement signatures of a tunable subset of excitations. We demonstrate the power of this approach by identifying a unique dynamical signature associated with the presence of an isolated Majorana zero mode in the postquench Hamiltonian, which results in a fractionalized jump in the entanglement entropy of the probe.
PHYSICAL REVIEW LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Bao Duong, Thin Nguyen
Summary: In this study, a novel method called LCIT (Latent representation-based Conditional Independence Test) is introduced for testing conditional independence based on representation learning. LCIT first learns to infer the latent representations of target variables X and Y that contain no information about conditioning variable Z, and then investigates the latent variables for any significant remaining dependencies using a conventional correlation test. LCIT outperforms several state-of-the-art baselines consistently and adapts well to both nonlinear, high-dimensional, and mixed data settings on a diverse collection of synthetic and real data sets.
KNOWLEDGE AND INFORMATION SYSTEMS
(2023)
