arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,153 records · Page 64Linked to original sources

arXiv2023

Scaling anisotropy with stationary background field in the near-Sun solar wind turbulence

The scaling of magnetic fluctuations provides crucial information for the understanding of solar wind turbulence. However, the observed magnetic fluctuations contain not only turbulence but also magnetic structures, leading to the violation of the time stationarity. This violation would conceal the true scaling and influence the determination of the sampling angle with respect to the local background magnetic field. Here, to investigate the scaling anisotropy, we utilize an easy but effective criterion $ϕ<10^\circ$ to ensure the time stationarity of the magnetic field, where $ϕ$ is the angle between the two averaged magnetic fields after cutting the interval into two halves. We study the scaling anisotropy using higher-order statistics of structure functions under the condition of stationarity for the near-Sun solar wind turbulence for the first time based on measurements obtained from Parker Solar Probe (PSP) at 0.17 au. We find that the scaling indices $ξ$ of magnetic field show a linear dependence on the order $p$ close to $ξ(p)=p/4$. The multifractal scaling of magnetic-trace structure functions becomes monoscaling close to $ξ(p)=p/3$ with the local magnetic field perpendicular to the sampling direction and close to $ξ(p)=p/4$ with the local magnetic field parallel to the sampling direction when measured with the stationary background magnetic field. The scaling of velocity-trace structure functions has similar but less significant changes. The near-Sun solar wind turbulence displays different scaling anisotropies with the near-Earth solar wind turbulence, suggesting the evolution of the nonlinear interaction process during the solar wind expansion.

physics.space-ph↗

arXiv2023

Early results from GLASS-JWST XVI: Discovering a bluer z~4-7 Universe through UV slopes

We use the GLASS-JWST Early Release Science NIRCam parallel observations to provide a first view of the UV continuum properties of NIRCam/F444W selected galaxies at 4 99% of the galaxies are blue star-forming galaxies with very low levels of dust (Avbeta~0.01+/-0.33). We find no statistically significant correlation for UV slope with redshift or UV magnitude. However, we find that in general galaxies at higher redshifts and fainter UV magnitudes have steeper UV slopes. We find a statistically significant correlation for UV slope with stellar mass, with galaxies with higher stellar mass showing shallower UV slopes. Individual fits to some of our galaxies reach the bluest UV slopes of beta~-3.1 allowed by stellar population models used in this analysis. Therefore, it is likely that stellar population models with higher amount of Lyman continuum leakage, AGN effects, and/or Population III contributions are required to accurately reproduce the rest-UV and optical properties of some of our bluest galaxies. This dust-free early view confirms that our current cosmological understanding of gradual mass + dust buildup of galaxies with cosmic time is largely accurate to describe the ~0.7-1.5 Gyr age window of the Universe. The abundance of a large population of UV faint dust-poor systems may point to a dominance of low-mass galaxies at z>6 playing a vital role in cosmic reionization.

astro-ph.GA↗

arXiv2023

Reproducible and Portable Big Data Analytics in the Cloud

Cloud computing has become a major approach to help reproduce computational experiments. Yet there are still two main difficulties in reproducing batch based big data analytics (including descriptive and predictive analytics) in the cloud. The first is how to automate end-to-end scalable execution of analytics including distributed environment provisioning, analytics pipeline description, parallel execution, and resource termination. The second is that an application developed for one cloud is difficult to be reproduced in another cloud, a.k.a. vendor lock-in problem. To tackle these problems, we leverage serverless computing and containerization techniques for automated scalable execution and reproducibility, and utilize the adapter design pattern to enable application portability and reproducibility across different clouds. We propose and develop an open-source toolkit that supports 1) fully automated end-to-end execution and reproduction via a single command, 2) automated data and configuration storage for each execution, 3) flexible client modes based on user preferences, 4) execution history query, and 5) simple reproduction of existing executions in the same environment or a different environment. We did extensive experiments on both AWS and Azure using four big data analytics applications that run on virtual CPU/GPU clusters. The experiments show our toolkit can achieve good execution performance, scalability, and efficient reproducibility for cloud-based big data analytics.

cs.DC↗

arXiv2023

Optimal quantum dataset for learning a unitary transformation

Unitary transformations formulate the time evolution of quantum states. How to learn a unitary transformation efficiently is a fundamental problem in quantum machine learning. The most natural and leading strategy is to train a quantum machine learning model based on a quantum dataset. Although the presence of more training data results in better models, using too much data reduces the efficiency of training. In this work, we solve the problem on the minimum size of sufficient quantum datasets for learning a unitary transformation exactly, which reveals the power and limitation of quantum data. First, we prove that the minimum size of a dataset with pure states is $2^n$ for learning an $n$-qubit unitary transformation. To fully explore the capability of quantum data, we introduce a practical quantum dataset consisting of $n+1$ elementary tensor product states that are sufficient for exact training. The main idea is to simplify the structure utilizing decoupling, which leads to an exponential improvement in the size of the datasets with pure states. Furthermore, we show that the size of the quantum dataset with mixed states can be reduced to a constant, which yields an optimal quantum dataset for learning a unitary. We showcase the applications of our results in oracle compiling and Hamiltonian simulation. Notably, to accurately simulate a 3-qubit one-dimensional nearest-neighbor Heisenberg model, our circuit only uses $96$ elementary quantum gates, which is significantly less than $4080$ gates in the circuit constructed by the Trotter-Suzuki product formula.

quant-ph↗

arXiv2023

Lower bound for the T count via unitary stabilizer nullity

We introduce magic measures to quantify the nonstabilizerness of multiqubit quantum gates and establish lower bounds on the $T$ count for fault-tolerant quantum computation. First, we introduce the stabilizer nullity of multi-qubit unitary, which is based on the subgroup of the quotient Pauli group associated with the unitary. This unitary stabilizer nullity extends the state-stabilizer nullity by Beverland et al. to a dynamic version. In particular, we show this nonstabilizerness measure has desirable properties such as subadditivity under composition and additivity under tensor product. Second, we prove that a given unitary's stabilizer nullity is a lower bound for the $T$ count, utilizing the above properties in gate synthesis. Third, we compare the state- and the unitary-stabilizer nullity, proving that the lower bounds for the $T$ count obtained by the unitary-stabilizer nullity are never less than the state-stabilizer nullity. Moreover, we show an explicit $n$-qubit unitary family of unitary-stabilizer nullity $2n$, which implies that its $T$ count is at least $2n$. This gives an example where the bounds derived by the unitary-stabilizer nullity strictly outperform the state-stabilizer nullity by a factor of $2$. We finally showcase the advantages of unitary-stabilizer nullity in estimating the $T$ count of quantum gates with interests.

quant-ph↗

arXiv2023

RMBench: Benchmarking Deep Reinforcement Learning for Robotic Manipulator Control

Reinforcement learning is applied to solve actual complex tasks from high-dimensional, sensory inputs. The last decade has developed a long list of reinforcement learning algorithms. Recent progress benefits from deep learning for raw sensory signal representation. One question naturally arises: how well do they perform concerning different robotic manipulation tasks? Benchmarks use objective performance metrics to offer a scientific way to compare algorithms. In this paper, we present RMBench, the first benchmark for robotic manipulations, which have high-dimensional continuous action and state spaces. We implement and evaluate reinforcement learning algorithms that directly use observed pixels as inputs. We report their average performance and learning curves to show their performance and stability of training. Our study concludes that none of the studied algorithms can handle all tasks well, soft Actor-Critic outperforms most algorithms in average reward and stability, and an algorithm combined with data augmentation may facilitate learning policies. Our code is publicly available at https://github.com/xiangyanfei212/RMBench-2022, including all benchmark tasks and studied algorithms.

cs.RO↗

arXiv2023

Autler-Townes splitting in the trap-loss fluorescence spectroscopy due to single-step direct Rydberg excitation of cesium cold atomic ensemble

We experimentally investigate trap-loss spectra of the cesium 6S1/2(F=4)-71P3/2 Rydberg state by combining the cesium atomic magneto-optical trap with the narrow-linewidth, continuously-tunable 318.6 nm ultraviolet laser. That is, the atoms in the magneto-optical trap are excited to the Rydberg state due to the ultraviolet laser single-step Rydberg excitation, which leads to the reduction of atomic fluorescence. Based on the trap-loss spectroscopy technology, the Autler-Townes splitting due to strong cooling laser is observed, and the parameter dependence of the AT splitting interval of trap-loss spectroscopy is investigated. Furthermore, the effective temperature of cold atoms is measured by means of simplified time-of-flight fluorescence imaging. In addition, closed-loop positive feedback power stabilization of 318.6 nm ultraviolet laser is carried out. This lays the foundation for further experimental research related to Rydberg atoms using ultraviolet lasers, which is of great significance for the development of quantum computing and quantum information fields.

physics.atom-ph↗

arXiv2023

Selectively Hard Negative Mining for Alleviating Gradient Vanishing in Image-Text Matching

Recently, a series of Image-Text Matching (ITM) methods achieve impressive performance. However, we observe that most existing ITM models suffer from gradients vanishing at the beginning of training, which makes these models prone to falling into local minima. Most ITM models adopt triplet loss with Hard Negative mining (HN) as the optimization objective. We find that optimizing an ITM model using only the hard negative samples can easily lead to gradient vanishing. In this paper, we derive the condition under which the gradient vanishes during training. When the difference between the positive pair similarity and the negative pair similarity is close to 0, the gradients on both the image and text encoders will approach 0. To alleviate the gradient vanishing problem, we propose a Selectively Hard Negative Mining (SelHN) strategy, which chooses whether to mine hard negative samples according to the gradient vanishing condition. SelHN can be plug-and-play applied to existing ITM models to give them better training behavior. To further ensure the back-propagation of gradients, we construct a Residual Visual Semantic Embedding model with SelHN, denoted as RVSE++. Extensive experiments on two ITM benchmarks demonstrate the strength of RVSE++, achieving state-of-the-art performance.

cs.CV↗

arXiv2023

Spoofed training data for speech spoofing countermeasure can be efficiently created using neural vocoders

A good training set for speech spoofing countermeasures requires diverse TTS and VC spoofing attacks, but generating TTS and VC spoofed trials for a target speaker may be technically demanding. Instead of using full-fledged TTS and VC systems, this study uses neural-network-based vocoders to do copy-synthesis on bona fide utterances. The output data can be used as spoofed data. To make better use of pairs of bona fide and spoofed data, this study introduces a contrastive feature loss that can be plugged into the standard training criterion. On the basis of the bona fide trials from the ASVspoof 2019 logical access training set, this study empirically compared a few training sets created in the proposed manner using a few neural non-autoregressive vocoders. Results on multiple test sets suggest good practices such as fine-tuning neural vocoders using bona fide data from the target domain. The results also demonstrated the effectiveness of the contrastive feature loss. Combining the best practices, the trained CM achieved overall competitive performance. Its EERs on the ASVspoof 2021 hidden subsets also outperformed the top-1 challenge submission.

eess.AS↗

arXiv2023

Unsupervised Deep Learning for IoT Time Series

IoT time series analysis has found numerous applications in a wide variety of areas, ranging from health informatics to network security. Nevertheless, the complex spatial temporal dynamics and high dimensionality of IoT time series make the analysis increasingly challenging. In recent years, the powerful feature extraction and representation learning capabilities of deep learning (DL) have provided an effective means for IoT time series analysis. However, few existing surveys on time series have systematically discussed unsupervised DL-based methods. To fill this void, we investigate unsupervised deep learning for IoT time series, i.e., unsupervised anomaly detection and clustering, under a unified framework. We also discuss the application scenarios, public datasets, existing challenges, and future research directions in this area.

cs.LG↗

arXiv2023

Chiral SQUID-metamaterial waveguide for circuit-QED

Superconducting metamaterials, which are designed and fabricated with structured fundamental circuit elements, have motivated recent developments of exploring unconventional quantum phenomena in circuit quantum electrodynamics (circuit-QED). We propose a method to engineer 1D Josephson metamaterial as a chiral waveguide by considering a programmed spatiotemporal modulation on its effective impedance. The modulation currents are in the form of traveling waves which phase velocities are much slower than the propagation speed of microwave photons. Due to the Brillouin-scattering process, non-trivial spectrum regimes where photons can propagate unidirectionally emerge. Considering superconducting qubits coupling with this metamaterial waveguide, we analyze both Markovian and non-Markovian quantum dynamics, and find that superconducting qubits can dissipate photons unidirectionally. Moreover, we show that our proposal can be extended a cascaded quantum network with multiple nodes, where chiral photon transport between remote qubits can be realized. Our work might open the possibilities to exploit SQUID metamaterials for realizing unidirectional photon transport in circuit-QED platforms.

quant-ph↗

arXiv2023

RIS-Assisted Jamming Rejection and Path Planning for UAV-Borne IoT Platform: A New Deep Reinforcement Learning Framework

This paper presents a new deep reinforcement learning (DRL)-based approach to the trajectory planning and jamming rejection of an unmanned aerial vehicle (UAV) for the Internet-of-Things (IoT) applications. Jamming can prevent timely delivery of sensing data and reception of operation instructions. With the assistance of a reconfigurable intelligent surface (RIS), we propose to augment the radio environment, suppress jamming signals, and enhance the desired signals. The UAV is designed to learn its trajectory and the RIS configuration based solely on changes in its received data rate, using the latest deep deterministic policy gradient (DDPG) and twin delayed DDPG (TD3) models. Simulations show that the proposed DRL algorithms give the UAV with strong resistance against jamming and that the TD3 algorithm exhibits faster and smoother convergence than the DDPG algorithm, and suits better for larger RISs. This DRL-based approach eliminates the need for knowledge of the channels involving the RIS and jammer, thereby offering significant practical value.

eess.SP↗

arXiv2023

IMPORTANT-Net: Integrated MRI Multi-Parameter Reinforcement Fusion Generator with Attention Network for Synthesizing Absent Data

Magnetic resonance imaging (MRI) is highly sensitive for lesion detection in the breasts. Sequences obtained with different settings can capture the specific characteristics of lesions. Such multi-parameter MRI information has been shown to improve radiologist performance in lesion classification, as well as improving the performance of artificial intelligence models in various tasks. However, obtaining multi-parameter MRI makes the examination costly in both financial and time perspectives, and there may be safety concerns for special populations, thus making acquisition of the full spectrum of MRI sequences less durable. In this study, different than naive input fusion or feature concatenation from existing MRI parameters, a novel $\textbf{I}$ntegrated MRI $\textbf{M}$ulti-$\textbf{P}$arameter reinf$\textbf{O}$rcement fusion generato$\textbf{R}$ wi$\textbf{T}$h $\textbf{A}$tte$\textbf{NT}$ion Network (IMPORTANT-Net) is developed to generate missing parameters. First, the parameter reconstruction module is used to encode and restore the existing MRI parameters to obtain the corresponding latent representation information at any scale level. Then the multi-parameter fusion with attention module enables the interaction of the encoded information from different parameters through a set of algorithmic strategies, and applies different weights to the information through the attention mechanism after information fusion to obtain refined representation information. Finally, a reinforcement fusion scheme embedded in a $V^{-}$-shape generation module is used to combine the hierarchical representations to generate the missing MRI parameter. Results showed that our IMPORTANT-Net is capable of generating missing MRI parameters and outperforms comparable state-of-the-art networks. Our code is available at https://github.com/Netherlands-Cancer-Institute/MRI_IMPORTANT_NET.

eess.IV↗

arXiv2023

Synthesis-based Imaging-Differentiation Representation Learning for Multi-Sequence 3D/4D MRI

Multi-sequence MRIs can be necessary for reliable diagnosis in clinical practice due to the complimentary information within sequences. However, redundant information exists across sequences, which interferes with mining efficient representations by modern machine learning or deep learning models. To handle various clinical scenarios, we propose a sequence-to-sequence generation framework (Seq2Seq) for imaging-differentiation representation learning. In this study, not only do we propose arbitrary 3D/4D sequence generation within one model to generate any specified target sequence, but also we are able to rank the importance of each sequence based on a new metric estimating the difficulty of a sequence being generated. Furthermore, we also exploit the generation inability of the model to extract regions that contain unique information for each sequence. We conduct extensive experiments using three datasets including a toy dataset of 20,000 simulated subjects, a brain MRI dataset of 1,251 subjects, and a breast MRI dataset of 2,101 subjects, to demonstrate that (1) our proposed Seq2Seq is efficient and lightweight for complex clinical datasets and can achieve excellent image quality; (2) top-ranking sequences can be used to replace complete sequences with non-inferior performance; (3) combining MRI with our imaging-differentiation map leads to better performance in clinical tasks such as glioblastoma MGMT promoter methylation status prediction and breast cancer pathological complete response status prediction. Our code is available at https://github.com/fiy2W/mri_seq2seq.

cs.CV↗

arXiv2023

The Power of External Memory in Increasing Predictive Model Capacity

One way of introducing sparsity into deep networks is by attaching an external table of parameters that is sparsely looked up at different layers of the network. By storing the bulk of the parameters in the external table, one can increase the capacity of the model without necessarily increasing the inference time. Two crucial questions in this setting are then: what is the lookup function for accessing the table and how are the contents of the table consumed? Prominent methods for accessing the table include 1) using words/wordpieces token-ids as table indices, 2) LSH hashing the token vector in each layer into a table of buckets, and 3) learnable softmax style routing to a table entry. The ways to consume the contents include adding/concatenating to input representation, and using the contents as expert networks that specialize to different inputs. In this work, we conduct rigorous experimental evaluations of existing ideas and their combinations. We also introduce a new method, alternating updates, that enables access to an increased token dimension without increasing the computation time, and demonstrate its effectiveness in language modeling.

cs.LG↗

arXiv2023

The PartialSpoof Database and Countermeasures for the Detection of Short Fake Speech Segments Embedded in an Utterance

Automatic speaker verification is susceptible to various manipulations and spoofing, such as text-to-speech synthesis, voice conversion, replay, tampering, adversarial attacks, and so on. We consider a new spoofing scenario called "Partial Spoof" (PS) in which synthesized or transformed speech segments are embedded into a bona fide utterance. While existing countermeasures (CMs) can detect fully spoofed utterances, there is a need for their adaptation or extension to the PS scenario. We propose various improvements to construct a significantly more accurate CM that can detect and locate short-generated spoofed speech segments at finer temporal resolutions. First, we introduce newly developed self-supervised pre-trained models as enhanced feature extractors. Second, we extend our PartialSpoof database by adding segment labels for various temporal resolutions. Since the short spoofed speech segments to be embedded by attackers are of variable length, six different temporal resolutions are considered, ranging from as short as 20 ms to as large as 640 ms. Third, we propose a new CM that enables the simultaneous use of the segment-level labels at different temporal resolutions as well as utterance-level labels to execute utterance- and segment-level detection at the same time. We also show that the proposed CM is capable of detecting spoofing at the utterance level with low error rates in the PS scenario as well as in a related logical access (LA) scenario. The equal error rates of utterance-level detection on the PartialSpoof database and ASVspoof 2019 LA database were 0.77 and 0.90%, respectively.

eess.AS↗

arXiv2023

Exact entanglement cost of quantum states and channels under PPT-preserving operations

This paper establishes single-letter formulas for the exact entanglement cost of simulating quantum channels under free quantum operations that completely preserve positivity of the partial transpose (PPT). First, we introduce the $κ$-entanglement measure for point-to-point quantum channels, based on the idea of the $κ$-entanglement of bipartite states, and we establish several fundamental properties for it, including amortization collapse, monotonicity under PPT superchannels, additivity, normalization, faithfulness, and non-convexity. Second, we introduce and solve the exact entanglement cost for simulating quantum channels in both the parallel and sequential settings, along with the assistance of free PPT-preserving operations. In particular, we establish that the entanglement cost in both cases is given by the same single-letter formula, the $κ$-entanglement measure of a quantum channel. We further show that this cost is equal to the largest $κ$-entanglement that can be shared or generated by the sender and receiver of the channel. This formula is calculable by a semidefinite program, thus allowing for an efficiently computable solution for general quantum channels. Noting that the sequential regime is more powerful than the parallel regime, another notable implication of our result is that both regimes have the same power for exact quantum channel simulation, when PPT superchannels are free. For several basic Gaussian quantum channels, we show that the exact entanglement cost is given by the Holevo--Werner formula [Holevo and Werner, Phys. Rev. A 63, 032312 (2001)], giving an operational meaning of the Holevo-Werner quantity for these channels.

quant-ph↗

arXiv2023

Attacking Important Pixels for Anchor-free Detectors

Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbation can completely change the prediction result. Existing adversarial attacks on object detection focus on attacking anchor-based detectors, which may not work well for anchor-free detectors. In this paper, we propose the first adversarial attack dedicated to anchor-free detectors. It is a category-wise attack that attacks important pixels of all instances of a category simultaneously. Our attack manifests in two forms, sparse category-wise attack (SCA) and dense category-wise attack (DCA), that minimize the $L_0$ and $L_\infty$ norm-based perturbations, respectively. For DCA, we present three variants, DCA-G, DCA-L, and DCA-S, that select a global region, a local region, and a semantic region, respectively, to attack. Our experiments on large-scale benchmark datasets including PascalVOC, MS-COCO, and MS-COCO Keypoints indicate that our proposed methods achieve state-of-the-art attack performance and transferability on both object detection and human pose estimation tasks.

cs.CV↗