arXiv subjects

Zhe Xu

Publications and source records attributed to Zhe Xu.

At least 163 records · Page 9Linked to original sources

arXiv2022

Statistical analysis of circular-ribbon flares

Circular-ribbon flares (CFs) are a special type of solar flares owing to their particular magnetic topology. In this paper, we conducted a comprehensive statistical analysis of 134 CFs from 2011 September to 2017 June, including four B-class, 82 C-class, 40 M-class, and eight X-class flares, respectively. The flares were observed by the Atmospheric Imaging Assembly (AIA) on board the Solar Dynamics Observatory (SDO) spacecraft. The physical properties of CFs are derived, including the location, area ($A_{CF}$), equivalent radius ($r_{CF}$) assuming a semi-spherical fan dome, lifetime ($τ_{CF}$), and peak SXR flux in 1$-$8 Å. It is found that all CFs are located in active regions, with the latitudes between -30$^\circ$ and 30$^\circ$. The distributions of areas and lifetimes could be fitted with a log-normal function. There is a positive correlation between the lifetime and area. The peak SXR flux in 1$-$8 Å is well in accord with a power-law distribution with an index of $-$1.42. For the 134 CFs, 57\% of them are accompanied by remote brightenings or ribbons. A positive correlation exists between the total length ($L_{RB}$) and average distance ($D_{RB}$) of remote brightenings. About 47\% and 51\% of the 134 CFs are related to type III radio bursts and jets, respectively. The association rates are independent of flare energies. About 38\% of CFs are related to mini-filament eruptions, and the association rates increase with flare classes. Only 28\% of CFs are related to CMEs, meaning that a majority of them are confined rather than eruptive events. There is a positive correlation between the CME speed and peak SXR flux in 1$-$8 Å, and faster CMEs tend to be wider.

astro-ph.SR↗

arXiv2022

All-Around Real Label Supervision: Cyclic Prototype Consistency Learning for Semi-supervised Medical Image Segmentation

Semi-supervised learning has substantially advanced medical image segmentation since it alleviates the heavy burden of acquiring the costly expert-examined annotations. Especially, the consistency-based approaches have attracted more attention for their superior performance, wherein the real labels are only utilized to supervise their paired images via supervised loss while the unlabeled images are exploited by enforcing the perturbation-based \textit{"unsupervised"} consistency without explicit guidance from those real labels. However, intuitively, the expert-examined real labels contain more reliable supervision signals. Observing this, we ask an unexplored but interesting question: can we exploit the unlabeled data via explicit real label supervision for semi-supervised training? To this end, we discard the previous perturbation-based consistency but absorb the essence of non-parametric prototype learning. Based on the prototypical network, we then propose a novel cyclic prototype consistency learning (CPCL) framework, which is constructed by a labeled-to-unlabeled (L2U) prototypical forward process and an unlabeled-to-labeled (U2L) backward process. Such two processes synergistically enhance the segmentation network by encouraging more discriminative and compact features. In this way, our framework turns previous \textit{"unsupervised"} consistency into new \textit{"supervised"} consistency, obtaining the \textit{"all-around real label supervision"} property of our method. Extensive experiments on brain tumor segmentation from MRI and kidney segmentation from CT images show that our CPCL can effectively exploit the unlabeled data and outperform other state-of-the-art semi-supervised medical image segmentation methods.

eess.IV↗

arXiv2022

Double-Uncertainty Guided Spatial and Temporal Consistency Regularization Weighting for Learning-based Abdominal Registration

In order to tackle the difficulty associated with the ill-posed nature of the image registration problem, regularization is often used to constrain the solution space. For most learning-based registration approaches, the regularization usually has a fixed weight and only constrains the spatial transformation. Such convention has two limitations: (i) Besides the laborious grid search for the optimal fixed weight, the regularization strength of a specific image pair should be associated with the content of the images, thus the "one value fits all" training scheme is not ideal; (ii) Only spatially regularizing the transformation may neglect some informative clues related to the ill-posedness. In this study, we propose a mean-teacher based registration framework, which incorporates an additional temporal consistency regularization term by encouraging the teacher model's prediction to be consistent with that of the student model. More importantly, instead of searching for a fixed weight, the teacher enables automatically adjusting the weights of the spatial regularization and the temporal consistency regularization by taking advantage of the transformation uncertainty and appearance uncertainty. Extensive experiments on the challenging abdominal CT-MRI registration show that our training strategy can promisingly advance the original learning-based method in terms of efficient hyperparameter tuning and a better tradeoff between accuracy and smoothness.

cs.CV↗

arXiv2022

Magnetic field induced hair structure in charmonium gluon-dissociation

We study electromagnetic field effect on charmonium gluon-dissociation in quark-gluon plasma. With the effective Hamiltonian derived from QCD multipole expansion under an external electromagnetic field, we first solve the two-body Schrödinger equation for a pair of charm quarks with mean field potentials for color and electromagnetic interactions and obtain the charmonium binding energies and wave functions, and then calculate the gluon-dissociation cross-section and decay width by taking the color electric and magnetic dipole interactions as perturbations above the mean field and employing Fermi's Golden Rule. Considering the charmonium deformation in magnetic field, the discrete Landau energy levels make the dissociation cross-section grow hair, and the electric dipole channel is significantly changed, especially for the $P-$wave states $χ_{c0}$ and $χ_{c\pm}$. From our numerical calculation, the magnetic field strength $eB=5~m_π^2$ already changes the gluon dissociation strongly, which may indicate measurable effects in high-energy nuclear collisions.

nucl-th↗

arXiv2022

Joint Inference of Reward Machines and Policies for Reinforcement Learning

Incorporating high-level knowledge is an effective way to expedite reinforcement learning (RL), especially for complex tasks with sparse rewards. We investigate an RL problem where the high-level knowledge is in the form of reward machines, i.e., a type of Mealy machine that encodes the reward functions. We focus on a setting in which this knowledge is a priori not available to the learning agent. We develop an iterative algorithm that performs joint inference of reward machines and policies for RL (more specifically, q-learning). In each iteration, the algorithm maintains a hypothesis reward machine and a sample of RL episodes. It derives q-functions from the current hypothesis reward machine, and performs RL to update the q-functions. While performing RL, the algorithm updates the sample by adding RL episodes along which the obtained rewards are inconsistent with the rewards based on the current hypothesis reward machine. In the next iteration, the algorithm infers a new hypothesis reward machine from the updated sample. Based on an equivalence relationship we defined between states of reward machines, we transfer the q-functions between the hypothesis reward machines in consecutive iterations. We prove that the proposed algorithm converges almost surely to an optimal policy in the limit if a minimal reward machine can be inferred and the maximal length of each RL episode is sufficiently long. The experiments show that learning high-level knowledge in the form of reward machines can lead to fast convergence to optimal policies in RL, while standard RL methods such as q-learning and hierarchical RL methods fail to converge to optimal policies after a substantial number of training steps in many tasks.

cs.AI↗

arXiv2022

Trust It or Not: Confidence-Guided Automatic Radiology Report Generation

Medical imaging plays a pivotal role in diagnosis and treatment in clinical practice. Inspired by the significant progress in automatic image captioning, various deep learning (DL)-based methods have been proposed to generate radiology reports for medical images. Despite promising results, previous works overlook the uncertainties of their models and are thus unable to provide clinicians with the reliability/confidence of the generated radiology reports to assist their decision-making. In this paper, we propose a novel method to explicitly quantify both the visual uncertainty and the textual uncertainty for DL-based radiology report generation. Such multi-modal uncertainties can sufficiently capture the model confidence degree at both the report level and the sentence level, and thus they are further leveraged to weight the losses for more comprehensive model optimization. Experimental results have demonstrated that the proposed method for model uncertainty characterization and estimation can produce more reliable confidence scores for radiology report generation, and the modified loss function, which takes into account the uncertainties, leads to better model performance on two public radiology report datasets. In addition, the quality of the automatically generated reports was manually evaluated by human raters and the results also indicate that the proposed uncertainties can reflect the variance of clinical diagnosis.

cs.CV↗

arXiv2022

A partial filament eruption in three steps induced by external magnetic reconnection

We present an investigation of partial filament eruption on 2012 June 17 in the active region NOAA 11504. For the first time, we observed the vertical splitting process during the partial eruption with high resolution narrow band images at 10830 . The active filament was rooted in a small sunspot of the active region. Particularly, it underwent the partial eruption in three steps, i.e. the precursor, the first eruption, and the second eruption, while the later two were associated with a C1.0 flare and a C3.9 flare, respectively. During the precursor, slow magnetic reconnection took place between the filament and the adjoining loops that also rooted in the sunspot. The continuous reconnection not only caused the filament to split into three groups of threads vertically but also formed a new filament, which was growing and accompanied brightening took place around the site. Subsequently, the growing filament erupted together with one group splitted threads, resulted in the first eruption. At the beginning of the first eruption, a subsequent magnetic reconnection occurred between the erupting splitted threads and another ambient magnetic loop. After about three minutes, the second eruption occurred as a result of the eruption of two larger unstable filaments induced by the magnetic reconnection. The high-resolution observation provides a direct evidence that magnetic reconnection between filament and its ambient magnetic fields could induce the vertical splitting of the filament, resulting in partial eruption.

astro-ph.SR↗

arXiv2022

Non-Parametric Neuro-Adaptive Coordination of Multi-Agent Systems

We develop a learning-based algorithm for the distributed formation control of networked multi-agent systems governed by unknown, nonlinear dynamics. Most existing algorithms either assume certain parametric forms for the unknown dynamic terms or resort to unnecessarily large control inputs in order to provide theoretical guarantees. The proposed algorithm avoids these drawbacks by integrating neural network-based learning with adaptive control in a two-step procedure. In the first step of the algorithm, each agent learns a controller, represented as a neural network, using training data that correspond to a collection of formation tasks and agent parameters. These parameters and tasks are derived by varying the nominal agent parameters and the formation specifications of the task in hand, respectively. In the second step of the algorithm, each agent incorporates the trained neural network into an online and adaptive control policy in such a way that the behavior of the multi-agent closed-loop system satisfies a user-defined formation task. Both the learning phase and the adaptive control policy are distributed, in the sense that each agent computes its own actions using only local information from its neighboring agents. The proposed algorithm does not use any a priori information on the agents' unknown dynamic terms or any approximation schemes. We provide formal theoretical guarantees on the achievement of the formation task.

eess.SY↗

arXiv2022

Weighted Graph-Based Signal Temporal Logic Inference Using Neural Networks

Extracting spatial-temporal knowledge from data is useful in many applications. It is important that the obtained knowledge is human-interpretable and amenable to formal analysis. In this paper, we propose a method that trains neural networks to learn spatial-temporal properties in the form of weighted graph-based signal temporal logic (wGSTL) formulas. For learning wGSTL formulas, we introduce a flexible wGSTL formula structure in which the user's preference can be applied in the inferred wGSTL formulas. In the proposed framework, each neuron of the neural networks corresponds to a subformula in a flexible wGSTL formula structure. We initially train a neural network to learn the wGSTL operators and then train a second neural network to learn the parameters in a flexible wGSTL formula structure. We use a COVID-19 dataset and a rain prediction dataset to evaluate the performance of the proposed framework and algorithms. We compare the performance of the proposed framework with three baseline classification methods including K-nearest neighbors, decision trees, support vector machine, and artificial neural networks. The classification accuracy obtained by the proposed framework is comparable with the baseline classification methods.

cs.AI↗

arXiv2021

On the Detection of Markov Decision Processes

We study the detection problem for a finite set of Markov decision processes (MDPs) where the MDPs have the same state and action spaces but possibly different probabilistic transition functions. Any one of these MDPs could be the model for some underlying controlled stochastic process, but it is unknown a priori which MDP is the ground truth. We investigate whether it is possible to asymptotically detect the ground truth MDP model perfectly based on a single observed history (state-action sequence). Since the generation of histories depends on the policy adopted to control the MDPs, we discuss the existence and synthesis of policies that allow for perfect detection. We start with the case of two MDPs and establish a necessary and sufficient condition for the existence of policies that lead to perfect detection. Based on this condition, we then develop an algorithm that efficiently (in time polynomial in the size of the MDPs) determines the existence of policies and synthesizes one when they exist. We further extend the results to the more general case where there are more than two MDPs in the candidate set, and we develop a policy synthesis algorithm based on the breadth-first search and recursion. We demonstrate the effectiveness of our algorithms through numerical examples.

math.OC↗

arXiv2021

Graph Sanitation with Application to Node Classification

The past decades have witnessed the prosperity of graph mining, with a multitude of sophisticated models and algorithms designed for various mining tasks, such as ranking, classification, clustering and anomaly detection. Generally speaking, the vast majority of the existing works aim to answer the following question, that is, given a graph, what is the best way to mine it? In this paper, we introduce the graph sanitation problem, to answer an orthogonal question. That is, given a mining task and an initial graph, what is the best way to improve the initially provided graph? By learning a better graph as part of the input of the mining model, it is expected to benefit graph mining in a variety of settings, ranging from denoising, imputation to defense. We formulate the graph sanitation problem as a bilevel optimization problem, and further instantiate it by semi-supervised node classification, together with an effective solver named GaSoliNe. Extensive experimental results demonstrate that the proposed method is (1) broadly applicable with respect to different graph neural network models and flexible graph modification strategies, (2) effective in improving the node classification accuracy on both the original and contaminated graphs in various perturbation scenarios. In particular, it brings up to 25% performance improvement over the existing robust graph neural network methods.

cs.LG↗

arXiv2021

Formation of a Solar Filament by Magnetic Reconnection and Associated Chromospheric Evaporation and Subsequent Coronal Condensation

We present the first observation of a solar filament formed by magnetic reconnection and associated chromospheric evaporation and subsequent coronal condensation. Driven by shearing motion during flux emergence, a sequential tether-cutting reconnection processes occurred and resulted in an M1.3 confined flare accompany with the formation of a sigmoid structure. It is found that the flare had a conjugate compact footpoint brightenings, which correspond to the footpoints of the sigmoid. Furthermore, observational evidence of explosive evaporation is well diagnosed at the conjugate footpoint brightenings in the impulsive phase of the flare. After the flare, continuous cool condensations formed at about the middle section of the sigmoid and then moved in opposite directions along the sigmoid, eventually leading to the formation of the filament. These observations suggest that magnetic reconnection not only can form the magnetic field structure of the filament, but also heat their chromospheric footpoints during their formation and drive chromospheric evaporation. As a result, the heated chromospheric plasma may be evaporated into the magnetic field structure of the filament, where the accumulated hot plasma might suffer from thermal instability or thermal non-equilibrium, causing catastrophic cooling and coronal condensation to form the cool dense material of the filament. This observation lends strong support to the evaporative-condensation model and highlights the crucial role of magnetic reconnection in forming both the magnetic field structure and the cool dense material of filaments.

astro-ph.SR↗

arXiv2021

Splitting of elliptic flow in non-central relativistic heavy-ion collisions

We predict a new effect due to the presence of the global vorticity in non-central relativistic heavy-ion collisions, namely a splitting of the elliptic flow parameter $v_2$ at non-zero rapidity. The size of the splitting is proposed as a new observable that can be used to constrain the initial vortical configuration of the produced QCD matter in experiments. The new findings are demonstrated by numerical calculations employing the parton cascade model, Boltzmann Approach of MultiParton Scatterings (BAMPS), for non-central Au + Au collisions at $\sqrt{s_{NN}} = 200 \ GeV$.

hep-ph↗

arXiv2021

Non-Parametric Neuro-Adaptive Control Subject to Task Specifications

We develop a learning-based algorithm for the control of autonomous systems governed by unknown, nonlinear dynamics to satisfy user-specified spatio-temporal tasks expressed as signal temporal logic specifications. Most existing algorithms either assume certain parametric forms for the unknown dynamic terms or resort to unnecessarily large control inputs in order to provide theoretical guarantees. The proposed algorithm addresses these drawbacks by integrating neural-network-based learning with adaptive control. More specifically, the algorithm learns a controller, represented as a neural network, using training data that correspond to a collection of system parameters and tasks. These parameters and tasks are derived by varying the nominal parameters and the spatio-temporal constraints of the user-specified task, respectively. It then incorporates this neural network into an online closed-form adaptive control policy in such a way that the resulting behavior satisfies the user-defined task. The proposed algorithm does not use any a priori information on the unknown dynamic terms or any approximation schemes. We provide formal theoretical guarantees on the satisfaction of the task. Numerical experiments on a robotic manipulator and a unicycle robot demonstrate that the proposed algorithm guarantees the satisfaction of 50 user-defined tasks, and outperforms control policies that do not employ online adaptation or the neural-network controller. Finally, we show that the proposed algorithm achieves greater performance than standard reinforcement-learning algorithms in the pendulum benchmarking environment.

cs.RO↗

arXiv2021

Oscillations and mass-draining that lead to a sympathetic eruption of a quiescent filament

In this paper, we present a multi-wavelength analysis to mass-draining and oscillations in a large quiescent filament prior to its successful eruption on 2015 April 28. The eruption of a smaller filament that was parallel and in close, 350 proximity was observed to induce longitudinal oscillations and enhance mass-draining within the filament of interest. The longitudinal oscillation with an amplitude of 25 Mm and 23 km underwent no damping during its observable cycle. Subsequently the slightly enhanced draining may have excited a eruption behind the limb, leading to a feedback that further enhanced the draining and induced simultaneous oscillations within the filament of interest. We find significant damping for these simultaneous oscillations, where the transverse oscillations proceeded with the amplitudes of 15 Mm and 14 km, while the longitudinal oscillations involved a larger displacement and velocity amplitude (57 Mm, 43 km). The second grouping of oscillation lasted for 2 cycles and had the similar period of 2 hours. From this, the curvature radius and transverse magnetic field strength of the magnetic dips supporting the filaments can be estimated to be 355 Mm and 34 G. The mass-draining within the filament of interest lasted for 14 hours. The apparent velocity grew from 35 km to 85 km, with the transition being coincident with the occurrence of the oscillations. We conclude that two filament eruptions are sympathetic, i.e. the eruption of the quiescent filament was triggered by the eruption of the nearby smaller filament.

astro-ph.SR↗

arXiv2021

Multi-Trigger-Key: Towards Multi-Task Privacy Preserving In Deep Learning

Deep learning-based Multi-Task Classification (MTC) is widely used in applications like facial attributes and healthcare that warrant strong privacy guarantees. In this work, we aim to protect sensitive information in the inference phase of MTC and propose a novel Multi-Trigger-Key (MTK) framework to achieve the privacy-preserving objective. MTK associates each secured task in the multi-task dataset with a specifically designed trigger-key. The true information can be revealed by adding the trigger-key if the user is authorized. We obtain such an MTK model by training it with a newly generated training set. To address the information leakage malaise resulting from correlations among different tasks, we generalize the training process by incorporating an MTK decoupling process with a controllable trade-off between the protective efficacy and the model performance. Theoretical guarantees and experimental results demonstrate the effectiveness of the privacy protection without appreciable hindering on the model performance.

cs.LG↗

arXiv2021

Decentralized Graph-Based Multi-Agent Reinforcement Learning Using Reward Machines

In multi-agent reinforcement learning (MARL), it is challenging for a collection of agents to learn complex temporally extended tasks. The difficulties lie in computational complexity and how to learn the high-level ideas behind reward functions. We study the graph-based Markov Decision Process (MDP) where the dynamics of neighboring agents are coupled. We use a reward machine (RM) to encode each agent's task and expose reward function internal structures. RM has the capacity to describe high-level knowledge and encode non-Markovian reward functions. We propose a decentralized learning algorithm to tackle computational complexity, called decentralized graph-based reinforcement learning using reward machines (DGRM), that equips each agent with a localized policy, allowing agents to make decisions independently, based on the information available to the agents. DGRM uses the actor-critic structure, and we introduce the tabular Q-function for discrete state problems. We show that the dependency of Q-function on other agents decreases exponentially as the distance between them increases. Furthermore, the complexity of DGRM is related to the local information size of the largest $κ$-hop neighborhood, and DGRM can find an $O(ρ^{κ+1})$-approximation of a stationary point of the objective function. To further improve efficiency, we also propose the deep DGRM algorithm, using deep neural networks to approximate the Q-function and policy function to solve large-scale or continuous state problems. The effectiveness of the proposed DGRM algorithm is evaluated by two case studies, UAV package delivery and COVID-19 pandemic mitigation. Experimental results show that local information is sufficient for DGRM and agents can accomplish complex tasks with the help of RM. DGRM improves the global accumulated reward by 119% compared to the baseline in the case of COVID-19 pandemic mitigation.

cs.MA↗

arXiv2021

Recurrent coronal jets observed by SDO/AIA

In this paper, we carry out multiwavelength observations of three recurring jets on 2014 November 7. The jets originated from the same region at the edge of AR 12205 and propagated along the same coronal loop. The eruptions were generated by magnetic reconnection, which is evidenced by continuous magnetic cancellation at the jet base. The projected initial velocity of the jet2 is 402 km s. The accelerations in the ascending and descending phases of jet2 are not consistent, the former is considerably larger than the value of solar gravitational acceleration at the solar surface, while the latter is lower than solar gravitational acceleration. There are two possible candidates of extra forces acting on jet2 during its propagation. One is the downward gas pressure from jet1 when it falls back and meets with jet2. The other is the viscous drag from the surrounding plasma during the fast propagation of jet2. As a contrast, the accelerations of jet3 in the rising and falling phases are constant, implying that the propagation of jet3 is not significantly influenced byextra forces.

astro-ph.SR↗