arXiv subjects

Yi Zheng

Publications and source records attributed to Yi Zheng.

At least 109 records · Page 6Linked to original sources

arXiv2024

Evolution equation for quantum coherence

Quantum coherence plays an important role in quantum resource theory, which is strongly related with entanglement. Similar to the entanglement evolution equation, we find the coherence evolution equation of quantum states through fully and strictly incoherent operation (FSIO) channels. In order to quantify the full coherence of qudit states, we define G-coherence and convex roof of G-coherence, and prove that the G-coherence is a strong coherence monotone and the convex roof of G-coherence is a coherence measure under FSIO, respectively. Furthermore, we prove a coherence evolution equation for arbitrary $d$-dimensional quantum pure and mixed states under FSIO channels, which generalizes the entanglement evolution equation for bipartite pure states. Our results will play an important role in the simplification of dynamical coherence measure.

quant-ph↗

arXiv2023

Critical assessment of water enthalpy characterization through dark environment evaporation

Comparative evaporation rate testing in the absence of solar irradiation is a widely adopted method for establishing and characterizing a reduced vaporization enthalpy of water within an interfacial solar evaporator. However, the assumption of equal energy input between cases is not generally valid, and renders this characterization method misleading. Larger evaporation rates result from interfacial evaporators in dark conditions, mostly due to expanded surface areas. This causes greater evaporation rates, more evaporative cooling, and larger temperature differentials with the environment. We provide theoretical and experimental evidence to prove that these temperature differences cause large differences in environmental energy input, such that equal energy input cannot be assumed. Evaluation of our data within a transient analytical model enables quantification of the energy input differences and calculation of the vaporization enthalpy. This shows that differences in evaporation rate correspond to differences in energy input, and that vaporization enthalpy is not reduced below the theoretical value. The magnitude of increased energy input also agrees well with the increase of dark environment evaporation rates typically reported for interfacial evaporators, providing an alternative explanation for previous literature results. Further, we exemplify that the results of this characterization method contradict differential scanning calorimetry results, which are often used to make conclusions about vaporization enthalpy reduction and water state modification within evaporator materials. Thus, we conclude that this method should not be used to make conclusions about vaporization enthalpy reductions within interfacial evaporator materials. These results emphasize that the current understanding of vaporization enthalpy reduction within evaporator materials requires re-evaluation.

physics.app-ph↗

arXiv2023

Differentiable modeling to unify machine learning and physical models and advance Geosciences

Process-Based Modeling (PBM) and Machine Learning (ML) are often perceived as distinct paradigms in the geosciences. Here we present differentiable geoscientific modeling as a powerful pathway toward dissolving the perceived barrier between them and ushering in a paradigm shift. For decades, PBM offered benefits in interpretability and physical consistency but struggled to efficiently leverage large datasets. ML methods, especially deep networks, presented strong predictive skills yet lacked the ability to answer specific scientific questions. While various methods have been proposed for ML-physics integration, an important underlying theme -- differentiable modeling -- is not sufficiently recognized. Here we outline the concepts, applicability, and significance of differentiable geoscientific modeling (DG). "Differentiable" refers to accurately and efficiently calculating gradients with respect to model variables, critically enabling the learning of high-dimensional unknown relationships. DG refers to a range of methods connecting varying amounts of prior knowledge to neural networks and training them together, capturing a different scope than physics-guided machine learning and emphasizing first principles. Preliminary evidence suggests DG offers better interpretability and causality than ML, improved generalizability and extrapolation capability, and strong potential for knowledge discovery, while approaching the performance of purely data-driven ML. DG models require less training data while scaling favorably in performance and efficiency with increasing amounts of data. With DG, geoscientists may be better able to frame and investigate questions, test hypotheses, and discover unrecognized linkages.

cs.LG↗

arXiv2023

Agents meet OKR: An Object and Key Results Driven Agent System with Hierarchical Self-Collaboration and Self-Evaluation

In this study, we introduce the concept of OKR-Agent designed to enhance the capabilities of Large Language Models (LLMs) in task-solving. Our approach utilizes both self-collaboration and self-correction mechanism, facilitated by hierarchical agents, to address the inherent complexities in task-solving. Our key observations are two-fold: first, effective task-solving demands in-depth domain knowledge and intricate reasoning, for which deploying specialized agents for individual sub-tasks can markedly enhance LLM performance. Second, task-solving intrinsically adheres to a hierarchical execution structure, comprising both high-level strategic planning and detailed task execution. Towards this end, our OKR-Agent paradigm aligns closely with this hierarchical structure, promising enhanced efficacy and adaptability across a range of scenarios. Specifically, our framework includes two novel modules: hierarchical Objects and Key Results generation and multi-level evaluation, each contributing to more efficient and robust task-solving. In practical, hierarchical OKR generation decomposes Objects into multiple sub-Objects and assigns new agents based on key results and agent responsibilities. These agents subsequently elaborate on their designated tasks and may further decompose them as necessary. Such generation operates recursively and hierarchically, culminating in a comprehensive set of detailed solutions. The multi-level evaluation module of OKR-Agent refines solution by leveraging feedback from all associated agents, optimizing each step of the process. This ensures solution is accurate, practical, and effectively address intricate task requirements, enhancing the overall reliability and quality of the outcome. Experimental results also show our method outperforms the previous methods on several tasks. Code and demo are available at https://okr-agent.github.io/

cs.CV↗

arXiv2023

High precision accelerator for our hybrid model of the redshift space power spectrum

Upcoming Large Scale Structure surveys aim to achieve an unprecedented level of precision in measuring galaxy clustering. However, accurately modeling these statistics may require theoretical templates that go beyond second-order perturbation theory, especially for achieving precision at smaller scales. In our previous work, we introduced a hybrid model for the redshift space power spectrum of galaxies. This model combines second-order templates with N-body simulations to capture the influence of scale-independent parameters on the galaxy power spectrum. However, the impact of scale-dependent parameters was addressed by precomputing a set of input statistics derived from computationally expensive N-body simulations. As a result, exploring the scale-dependent parameter space was not feasible in this approach. To address this challenge, we present an accelerated methodology that utilizes Gaussian processes, a machine learning technique, to emulate these input statistics. Our emulators exhibit remarkable accuracy, achieving reliable results with just 13 N-body simulations for training. We reproduce all necessary input statistics for a set of test simulations with an error of approximately 0.1 per cent in the parameter space within $5σ$ of the Planck predictions, specifically for scales around $k > 0.1$ $h$Mpc$^{-1}$. Following the training of our emulators, we can predict all inputs for our hybrid model in approximately 0.2,seconds at a specified redshift. Given that performing 13 N-body simulations is a manageable task, our present methodology enables us to construct efficient and highly accurate models of the galaxy power spectra within a manageable time frame.

astro-ph.CO↗

arXiv2023

Tunable non-additivity in Casimir-Lifshitz force between graphene gratings

We investigate the Casimir-Lifshitz force (CLF) between two identical graphene strip gratings, laid on finite dielectric substrates, by using the scattering matrix (S-matrix) approach derived from the Fourier Modal Method with Local Basis Functions (FMM-LBF). We fully take into account the high-order electromagnetic diffractions, the multiple scattering and the exact 2D feature of the graphene strips. We show that the non-additivity, which is one of the most interesting features of the CLF in general, is significantly high and can be modulated in situ, without any change in the actual material geometry and this by varying the graphene chemical potential. We discuss the nature of the geometrical effects and show the relevance of the geometric parameter d/D (i.e. the ratio between separation and grating period), which allows to explore the regions of parameters where the additive result is fully acceptable or where the full calculation is needed. This study can open to deeper experimental exploration of the non-additive features of the CLF with micro- or nano-electromechanical graphene-based systems.

cond-mat.mes-hall↗

arXiv2023

Fast generation of mock galaxy catalogues with COLA

We investigate the feasibility of using COmoving Lagrangian Acceleration (COLA) technique to efficiently generate galaxy mock catalogues that can accurately reproduce the statistical properties of observed galaxies. Our proposed scheme combines the subhalo abundance matching (SHAM) procedure with COLA simulations, utilizing only three free parameters: the scatter magnitude ($σ_{\rm scat}$) in SHAM, the initial redshift ($z_{\rm init}$) of the COLA simulation, and the time stride ($da$) used by COLA. In this proof-of-concept study, we focus on a subset of BOSS CMASS NGC galaxies within the redshift range $z\in [0.45, 0.55]$. We perform $\mathtt{GADGET}$ simulation and low-resolution COLA simulations with various combinations of $(z_{\rm init}, da)$, each using $1024^{3}$ particles in an $800~h^{-1}{\rm Mpc}$ box. By minimizing the difference between COLA mock and CMASS NGC galaxies for the monopole of the two-point correlation function (2PCF), we obtain the optimal $σ_{\rm scat}$. We have found that by setting $z_{\rm init}=29$ and $da=1/30$, we achieve a good agreement between COLA mock and CMASS NGC galaxies within the range of 4 to $20~h^{-1}{\rm Mpc}$, with a computational cost two orders of magnitude lower than that of the N-body code. Moreover, a detailed verification is performed by comparing various statistical properties, such as anisotropic 2PCF, three-point clustering, and power spectrum multipoles, which shows similar performance between GADGET mock and COLA mock catalogues with the CMASS NGC galaxies. Furthermore, we assess the robustness of the COLA mock catalogues across different cosmological models, demonstrating consistent results in the resulting 2PCFs. Our findings suggest that COLA simulations are a promising tool for efficiently generating mock catalogues for emulators and machine learning analyses in exploring the large-scale structure of the Universe.

astro-ph.CO↗

arXiv2023

Intelligent radiative thermostat induced by near-field radiative thermal diode

A radiative thermostat system senses its own temperature and automatically modulates heat transfer by turning on/off the cooling to maintain its temperature near a desired set point. Taking advantage of far- and near-field radiative thermal technologies, we propose an intelligent radiative thermostat induced by the combination of passive radiative cooling and near-field radiative thermal diode for thermal regulation at room temperature. The top passive radiative cooler in thermostat system with static thermal emissivity uses the cold outer space to passively cool itself all day, which can provide the bottom structure with the sub-ambient cold source. Meanwhile, using the phase-transition material vanadium dioxide, the bottom structure forms a near-field radiative thermal diode with the top cooler, which can significantly regulate the heat transfer between two terminals of the diode and then realize a stable temperature of the bottom structure. Besides, the backsided heat input of the thermostat has been taken into account according to real-world applications. Thermal performance of the proposed radiative thermostat design has been analyzed, showing that the coupling effect of static passive radiative cooling and dynamic internal heat transfer modulation can maintain an equilibrium temperature approximately locked within the phase transition region. Besides, after considering empirical indoor-to-outdoor heat flux, rendering its thermal performance closer to that of passive solar residential building walls, the calculation result proves that the radiative thermostat system can effectively modulate the temperature and stabilize it within a controllable range. Passive radiative thermostats driven by near-field radiative thermal diode can potentially enable intelligent temperature regulation technologies, for example, to moderate diurnal temperature in regions with extreme thermal swings.

physics.app-ph↗

arXiv2023

Gaussian beam quantum radar protocol

We present an entangled quantum radar protocol. It consists in scanning the sky with a thin Gaussian beam and measuring the travel time of the radiation reflected from the target, as in conventional radars. Here the Gaussian beam is composed of $N$ photons entangled in the frequency degrees of freedom. We show that this provides a $\sqrt{N}$ quantum enhancement over the unentangled case, as is usual in quantum metrology.

quant-ph↗

arXiv2023

Breaking the $τ_{\rm T}-f$ degeneracy of the kinetic Sunyaev Zel'dovich cosmology in redshift space

The 'optical depth-linear growth rate' ($τ_{\rm T}-f$) degeneracy is a long-standing problem in the kinetic Sunyaev Zel'dovich (kSZ) cosmology. It can be spontaneously broken in redshift space, where the velocity field leaves its own distinct imprint on the galaxies' redshift space positions and provides valuable information of the linear growth rate. We validate this idea with the Fisher matrix and Monte Carlo Markov Chain techniques in this work, finding that the level of this degeneracy breaking is further enhanced on non-linear scales due to the non-linear evolution of the density and velocity fields, if we have a good prior knowledge of the non-linear bias of galaxies. This result emphasizes the importance of the redshift space analysis of the kSZ effect and its potential as a powerful cosmological probe, especially on non-linear scales. As a by-product, we develop a non-linear model of the redshift space density-weighted pairwise kSZ power spectrum. The fitted $f$ and $τ_{\rm T}$ values from this model are shown to be accurate within $1-2σ$ ranges of the fiducial ones when confronted to the mock galaxies mimicking a DESI+CMB-S4 survey combination, even on small scales of $k\sim 0.5h/{\rm Mpc}$.

astro-ph.CO↗

arXiv2023

Experimental verification of a coherence factorization law for quantum states

As a quantum resource, quantum coherence plays an important role in modern physics. Many coherence measures and their relations with entanglement have been proposed, and the dynamics of entanglement has been experimentally studied. However, the knowledge of general results for coherence dynamics in open systems is limited. Here we propose a coherence factorization law, which describes the evolution of coherence passing through any noisy channels characterized by genuinely incoherent operations. We use photons to implement the quantum operations and experimentally verify the law for qubits and qutrits. Our work is a step toward the understanding of the evolution of coherence when the system interacts with the environment, and will boost the study of more general laws of coherence.

quant-ph↗

arXiv2023

AraMUS: Pushing the Limits of Data and Model Scale for Arabic Natural Language Processing

Developing monolingual large Pre-trained Language Models (PLMs) is shown to be very successful in handling different tasks in Natural Language Processing (NLP). In this work, we present AraMUS, the largest Arabic PLM with 11B parameters trained on 529GB of high-quality Arabic textual data. AraMUS achieves state-of-the-art performances on a diverse set of Arabic classification and generative tasks. Moreover, AraMUS shows impressive few-shot learning abilities compared with the best existing Arabic PLMs.

cs.CL↗

arXiv2023

Reconstructing the multiphoton spatial wave function with coincidence wavefront sensing

The quantum wave function of multiple particles provides additional information which is inaccessible to detectors working alone. Here, we introduce the coincidence wavefront sensing (CWS) method to reconstruct the phase of the multiphoton transverse spatial wave function. The spatially resolved coincidence photon counting is involved. Numerical simulations of two-photon cases using the weak measurement wavefront sensor are performed to test its correctness, and the phase information hidden in the correlation are revealed. Our work provides a direct spatial way to characterize multipartite quantum systems, and leads to fundamental studies like experimental Bohmian mechanics and applications in quantum optical technologies.

quant-ph↗

arXiv2023

Egalitarian ORAM: Wear-Leveling for ORAM

While non-volatile memories (NVMs) provide several desirable characteristics like better density and comparable energy efficiency than DRAM, DRAM-like performance, and disk-like durability, the limited endurance NVMs manifest remains a challenge with these memories. Indeed, the endurance constraints of NVMs can prevent solutions that are commonly employed for other mainstream memories like DRAM from being carried over as-is to NVMs. Specifically, in this work we observe that, Oblivious RAM (ORAM) primitive, the state-ofart solution to tackle memory bus side channel vulnerability, while widely studied for DRAMs, is particularly challenging to implement as-is for NVMs as it severely affects endurance of NVMs. This is so, as the inherent nature of ORAM primitive causes an order of magnitude increase in write traffic and furthermore, causes some regions of memory to be written far more often than others. This non-uniform write traffic as manifested by ORAM primitive stands to severely affect the lifetime of non-volatile memories (1% of baseline without ORAM) to even make it impractical to address this security vulnerability

cs.AR↗

arXiv2023

Unlocking Hidden Spins in Centrosymmetric SnSe2 by Vacancy-Controlled Spin-Orbit Scattering

Spin current generation and manipulation remain the key challenge of spintronics, in which relativistic spinorbit coupling (SOC) play a ubiquitous role. In this letter, we demonstrate that hidden Rashba spins in the non-magnetic, centrosymmetric lattice of multilayer SnSe2 can be efficiently activated by spin-orbit scattering introduced by Se vacancies. Via vacancy scattering, conduction electrons with hidden spin-momentum locked polarizations acquire out-of-plane magnetization components, which effectively break the chiral symmetry between the two Se sublattices of an SnSe2 monolayer when electron spins start precession in the strong built-in Rashba SOC field. The resulting spin separations are manifested in quantum transport as vacancy concentrationand temperature-dependent crossovers from weak antilocalization (WAL) to weak localization (WL), with the distinctive spin relaxation mechanism of the Dyakonov-Perel type. Our study shows the great potential of twodimensional systems with hidden-spin textures for spintronics.

cond-mat.mes-hall↗

arXiv2023

Realization of all-optical underdamped stochastic Stirling engine

We experimentally demonstrate a nano-scale stochastic Stirling heat engine operating in the underdamped regime. The setup involves an optically levitated silica particle that is subjected to a power-varying optical trap and periodically coupled to a cold/hot reservoir via switching on/off active feedback cooling. We conduct a systematic investigation of the engine's performance and find that both the output work and efficiency approach their theoretical limits under quasi-static conditions. Furthermore, we examine the dependence of the output work fluctuation on the cycle time and temperature difference between the hot and cold reservoirs. We observe that the distribution has a Gaussian profile in the quasi-static regime, whereas it becomes asymmetric and non-Gaussian as the cycle duration time decreases. This non-Gaussianity is qualitatively attributed to the strong correlation of the particle's position within a cycle in the non-equilibrium regime. Our experiments provide valuable insights into stochastic thermodynamics in the underdamped regime and open up new possibilities for the design of future nano-machines.

physics.optics↗

arXiv2022

Measuring cosmic filament spin with the kinetic Sunyaev-Zel'dovich effect

The spin of intergalactic filaments has been predicted from simulations, and supported by tentative evidence from redshift-space filament shapes in a galaxy redshift survey: generally, a filament is redshifted on one side of its axis, and blueshifted on the other. Here, we investigate whether filament spins could have a measurable kinetic Sunyaev-Zel'dovich (kSZ) signal, from CMB photons being scattered by moving ionised gas; this pure velocity information is complementary to filament redshift-space shapes. We propose to measure the kSZ dipole by combining galaxy redshift surveys with CMB experiments. We base our S/N analyses first on an existing filament catalogue, and its combination with Planck data. We then investigate the detectability of the kSZ dipole using the combination of DESI or SKA-2 with next-stage CMB experiments. We find that the gas halos of filament galaxies co-rotating with filaments induce a stronger kSZ dipole signal than that from the diffuse filamentary gas, but both signals seem too small to be detected in near-term surveys such as DESI+future CMB experiments. But the combination of SKA-2 with future CMB experiments could give a more than 10$σ$ detection. The gain comes mainly from an increased area overlap and an increased number of filaments, but also the low noise and high resolution in future CMB experiments are important to capture signals from filaments small on the sky. Successful detection of the signals may help to find the gravitomagnetic effect in large-scale structure and advance our understanding of baryons in the cosmic web.

astro-ph.CO↗

arXiv2022

ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks

Causal chain reasoning (CCR) is an essential ability for many decision-making AI systems, which requires the model to build reliable causal chains by connecting causal pairs. However, CCR suffers from two main transitive problems: threshold effect and scene drift. In other words, the causal pairs to be spliced may have a conflicting threshold boundary or scenario. To address these issues, we propose a novel Reliable Causal chain reasoning framework~(ReCo), which introduces exogenous variables to represent the threshold and scene factors of each causal pair within the causal chain, and estimates the threshold and scene contradictions across exogenous variables via structural causal recurrent neural networks~(SRNN). Experiments show that ReCo outperforms a series of strong baselines on both Chinese and English CCR datasets. Moreover, by injecting reliable causal chain knowledge distilled by ReCo, BERT can achieve better performances on four downstream causal-related tasks than BERT models enhanced by other kinds of knowledge.

cs.AI↗