arXiv subjects

Chen Zhang

Publications and source records attributed to Chen Zhang.

At least 523 records · Page 29Linked to original sources

arXiv2021

iVPF: Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression

It is nontrivial to store rapidly growing big data nowadays, which demands high-performance lossless compression techniques. Likelihood-based generative models have witnessed their success on lossless compression, where flow based models are desirable in allowing exact data likelihood optimisation with bijective mappings. However, common continuous flows are in contradiction with the discreteness of coding schemes, which requires either 1) imposing strict constraints on flow models that degrades the performance or 2) coding numerous bijective mapping errors which reduces the efficiency. In this paper, we investigate volume preserving flows for lossless compression and show that a bijective mapping without error is possible. We propose Numerical Invertible Volume Preserving Flow (iVPF) which is derived from the general volume preserving flows. By introducing novel computation algorithms on flow models, an exact bijective mapping is achieved without any numerical error. We also propose a lossless compression algorithm based on iVPF. Experiments on various datasets show that the algorithm based on iVPF achieves state-of-the-art compression ratio over lightweight compression algorithms.

cs.LG↗

arXiv2021

Conditional Variational Autoencoder for Learned Image Reconstruction

Learned image reconstruction techniques using deep neural networks have recently gained popularity, and have delivered promising empirical results. However, most approaches focus on one single recovery for each observation, and thus neglect the uncertainty information. In this work, we develop a novel computational framework that approximates the posterior distribution of the unknown image at each query observation. The proposed framework is very flexible: It handles implicit noise models and priors, it incorporates the data formation process (i.e., the forward operator), and the learned reconstructive properties are transferable between different datasets. Once the network is trained using the conditional variational autoencoder loss, it provides a computationally efficient sampler for the approximate posterior distribution via feed-forward propagation, and the summarizing statistics of the generated samples are used for both point-estimation and uncertainty quantification. We illustrate the proposed framework with extensive numerical experiments on positron emission tomography (with both moderate and low count levels) showing that the framework generates high-quality samples when compared with state-of-the-art methods.

cs.CV↗

arXiv2021

Transient Synchronization Stability Analysis of Wind Farms with MMC-HVDC Integration Under Offshore AC Grid Fault

The MMC-HVDC connected offshore wind farms (OWFs) could suffer short circuit fault (SCF), whereas their transient stability is not well analysed. In this paper, the mechanism of the loss of synchronization (LOS) of this system is analysed considering the whole system state from the fault-on to the post-fault, and the discussion on fault type and fault clearance is addressed as well. A stability index is proposed to quantify the transient synchronization stability (TSS) of the system, which is capable to not only estimate whether the wind turbine generators (WTGs) be able to get resynchronized with the offshore MMC after the fault is cleared, but also to evaluate the performance of stability improving methods as well. Finally, a scenario of six cases is tested on the PSCAD/EMTDC simulation platform, where the performances of four existing stability improving methods are thoroughly compared via both numerical simulation and the proposed stability index.

eess.SY↗

arXiv2021

Twist-diameter coupling drives DNA twist changes by salt and temperature

DNA deformations play crucial roles in many biological processes and material applications. During DNA deformation, DNA structural parameters often exhibit non-trivial and counterintuitive couplings, such as the twist-stretch and twist-bending couplings. Here, we reveal an unexpectedly strong negative twist-diameter coupling through the synergy of magnetic-tweezers experiments, atomistic molecular dynamics simulations, and theoretical calculations. In experiments, the DNA twist angle always increases with the concentration of NaCl, KCl, or RbCl. Our simulations quantitatively reproduce salt-induced twist changes and reveal the underlying physical mechanism: the reduction of DNA diameter under a high salt concentration leads to the increase in DNA twist angle through a strong negative twist-diameter coupling. The twist-diameter coupling is mediated by two dihedral angles in DNA structure and the coupling constant is 4.5 kBT/(deg nm) for one base-pair. Based on this coupling constant, we predict the temperature-dependence of DNA twist -0.0102 deg/K per bp, which agrees with our and previous experimental results. Our analysis suggests that the twist-diameter coupling is a common driving force for salt- and temperature-induced DNA twist changes.

physics.bio-ph↗

arXiv2021

More on the Weak Gravity Conjecture via Convexity of Charged Operators

The Weak Gravity Conjecture has recently been re-formulated in terms of a particle with non-negative self-binding energy. Because of the dual conformal field theory (CFT) formulation in the anti-de Sitter space the conformal dimension $Δ(Q)$ of the lowest-dimension operator with charge Q under some global U(1) symmetry must be a convex function of Q. This property has been conjectured to hold for any (unitary) conformal field theory and generalized to larger global symmetry groups. Here we refine and further test the convex charge conjecture via semiclassical computations for fixed charge sectors of different theories in different dimensions. We analyze the convexity properties of the leading and next-to-leading order terms stemming from the semiclassical computation, de facto, extending previous tests beyond the leading perturbative contributions and to arbitrary charges. In particular, the leading contribution is sufficient to test convexity in the semiclassical computations. We also consider intriguing cases in which the models feature a transition from real to complex conformal dimensions either as a function of the charge or number of matter fields. As a relevant example of the first kind, we investigate the $O(N)$ model in $4+ε$ dimensions. As an example of the second type we consider the $U(N)\times U(M)$ model in $4-ε$ dimensions. Both models display a rich dynamics where, by changing the number of matter fields and/or charge, one can achieve dramatically different physical regimes. We discover that whenever a complex conformal dimension appears, the real part satisfies the convexity property.

hep-th↗

arXiv2021

Revisiting Self-Training for Few-Shot Learning of Language Model

As unlabeled data carry rich task-relevant information, they are proven useful for few-shot learning of language model. The question is how to effectively make use of such data. In this work, we revisit the self-training technique for language model fine-tuning and present a state-of-the-art prompt-based few-shot learner, SFLM. Given two views of a text sample via weak and strong augmentation techniques, SFLM generates a pseudo label on the weakly augmented version. Then, the model predicts the same pseudo label when fine-tuned with the strongly augmented version. This simple approach is shown to outperform other state-of-the-art supervised and semi-supervised counterparts on six sentence classification and six sentence-pair classification benchmarking tasks. In addition, SFLM only relies on a few in-domain unlabeled data. We conduct a comprehensive analysis to demonstrate the robustness of our proposed approach under various settings, including augmentation techniques, model scale, and few-shot knowledge transfer across tasks.

cs.CL↗

arXiv2021

Improving Object Permanence using Agent Actions and Reasoning

Object permanence in psychology means knowing that objects still exist even if they are no longer visible. It is a crucial concept for robots to operate autonomously in uncontrolled environments. Existing approaches learn object permanence from low-level perception, but perform poorly on more complex scenarios, like when objects are contained and carried by others. Knowledge about manipulation actions performed on an object prior to its disappearance allows us to reason about its location, e.g., that the object has been placed in a carrier. In this paper we argue that object permanence can be improved when the robot uses knowledge about executed actions and describe an approach to infer hidden object states from agent actions. We show that considering agent actions not only improves rule-based reasoning models but also purely neural approaches, showing its general applicability. Then, we conduct quantitative experiments on a snitch localization task using a dataset of 1,371 synthesized videos, where we compare the performance of different object permanence models with and without action annotations. We demonstrate that models with action annotations can significantly increase performance of both neural and rule-based approaches. Finally, we evaluate the usability of our approach in real-world applications by conducting qualitative experiments with two Universal Robots (UR5 and UR16e) in both lab and industrial settings. The robots complete benchmark tasks for a gearbox assembly and demonstrate the object permanence capabilities with real sensor data in an industrial environment.

cs.RO↗

arXiv2021

Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics

In this paper, we characterize the noise of stochastic gradients and analyze the noise-induced dynamics during training deep neural networks by gradient-based optimizers. Specifically, we firstly show that the stochastic gradient noise possesses finite variance, and therefore the classical Central Limit Theorem (CLT) applies; this indicates that the gradient noise is asymptotically Gaussian. Such an asymptotic result validates the wide-accepted assumption of Gaussian noise. We clarify that the recently observed phenomenon of heavy tails within gradient noise may not be intrinsic properties, but the consequence of insufficient mini-batch size; the gradient noise, which is a sum of limited i.i.d. random variables, has not reached the asymptotic regime of CLT, thus deviates from Gaussian. We quantitatively measure the goodness of Gaussian approximation of the noise, which supports our conclusion. Secondly, we analyze the noise-induced dynamics of stochastic gradient descent using the Langevin equation, granting for momentum hyperparameter in the optimizer with a physical interpretation. We then proceed to demonstrate the existence of the steady-state distribution of stochastic gradient descent and approximate the distribution at a small learning rate.

cs.LG↗

arXiv2021

PDAugment: Data Augmentation by Pitch and Duration Adjustments for Automatic Lyrics Transcription

Automatic lyrics transcription (ALT), which can be regarded as automatic speech recognition (ASR) on singing voice, is an interesting and practical topic in academia and industry. ALT has not been well developed mainly due to the dearth of paired singing voice and lyrics datasets for model training. Considering that there is a large amount of ASR training data, a straightforward method is to leverage ASR data to enhance ALT training. However, the improvement is marginal when training the ALT system directly with ASR data, because of the gap between the singing voice and standard speech data which is rooted in music-specific acoustic characteristics in singing voice. In this paper, we propose PDAugment, a data augmentation method that adjusts pitch and duration of speech at syllable level under the guidance of music scores to help ALT training. Specifically, we adjust the pitch and duration of each syllable in natural speech to those of the corresponding note extracted from music scores, so as to narrow the gap between natural speech and singing voice. Experiments on DSing30 and Dali corpus show that the ALT system equipped with our PDAugment outperforms previous state-of-the-art systems by 5.9% and 18.1% WERs respectively, demonstrating the effectiveness of PDAugment for ALT.

eess.AS↗

arXiv2021

Unified Interacting Quark Matter and its Astrophysical Implications

We investigate interacting quark matter (IQM), including the perturbative QCD correction and color superconductivity, for both up-down quark matter ($ud$QM) and strange quark matter (SQM). We first derive an equation of state (EOS) unifying all cases by a simple reparametrization and rescaling, through which we manage to maximally reduce the number of degrees of freedom. We find, in contrast to the conventional EOS $p=1/3(ρ-4B_{\rm eff})$ for non-interacting quark matter, that taking the extreme strongly interacting limit on the unified IQM EOS gives $p=ρ-2B_{\rm eff}$, where $B_{\rm eff}$ is the effective bag constant. We employ the unified EOS to explore the properties of pure interacting quark stars (IQSs) composed of IQM. We describe how recent astrophysical observations, such as the pulsar-mass measurements, the NICER analysis, and the binary merger gravitational-wave events GW170817, GW190425, and GW190814, further constrain the parameter space. An upper bound for the maximum allowed mass of IQSs is found to be $M_{\rm TOV}\lesssim 3.23 M_{\odot}$. Our analysis indicates a new possibility that the currently observed compact stars, including the recently reported GW190814's secondary component ($M=2.59^{+0.08}_{-0.09}\, M_{\odot}$), can be quark stars composed of interacting quark matter.

astro-ph.HE↗

arXiv2021

Dirac semimetal PdTe2 temperature-dependent quasiparticle dynamics and electron-phonon coupling

Dirac semimetal PdTe2 single-crystal temperature-dependent ultrafast carrier and phonon dynamics were studied using ultrafast optical pump-probe spectroscopy. Two distinct carrier and coherent phonons relaxation processes were identified in the 5 K - 300 K range. Quantitative analysis revealed a fast relaxation process (τ_f) occurring on a subpicosecond time scale which originated from electron-phonon thermalization. This was followed by a slower relaxation process (τ_s) with a time scale of ~ 7-9.5 ps which originated from phonon-assisted electron-hole recombination. Two significant vibrational modes resolved at all measured temperatures and corresponded to Te atoms in-plane (E_g), and out-of-plane (A_1g), motion. As temperature increased both phonon modes softened markedly. A_1g mode frequency monotonically decreased as temperature increased. Its damping rate remained virtually unchanged. As expected, E_g decreased uniformly as temperatures rose. At temperatures above 80 K, there was insignificant change. Test results suggested that pure dephasing played an important role in the relaxation processes. PdTe2 phonon is thought responsible for its superconductive properties. Examining phonons behavior should improve the understanding of its complex superconductivity.

cond-mat.mtrl-sci↗

arXiv2021

Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

In this paper, we present a new verification style reading comprehension dataset named VGaokao from Chinese Language tests of Gaokao. Different from existing efforts, the new dataset is originally designed for native speakers' evaluation, thus requiring more advanced language understanding skills. To address the challenges in VGaokao, we propose a novel Extract-Integrate-Compete approach, which iteratively selects complementary evidence with a novel query updating mechanism and adaptively distills supportive evidence, followed by a pairwise competition to push models to learn the subtle difference among similar text pieces. Experiments show that our methods outperform various baselines on VGaokao with retrieved complementary evidence, while having the merits of efficiency and explainability. Our dataset and code are released for further research.

cs.CL↗

arXiv2021

Eden: A Unified Environment Framework for Booming Reinforcement Learning Algorithms

With AlphaGo defeats top human players, reinforcement learning(RL) algorithms have gradually become the code-base of building stronger artificial intelligence(AI). The RL algorithm design firstly needs to adapt to the specific environment, so the designed environment guides the rapid and profound development of RL algorithms. However, the existing environments, which can be divided into real world games and customized toy environments, have obvious shortcomings. For real world games, it is designed for human entertainment, and too much difficult for most of RL researchers. For customized toy environments, there is no widely accepted unified evaluation standard for all RL algorithms. Therefore, we introduce the first virtual user-friendly environment framework for RL. In this framework, the environment can be easily configured to realize all kinds of RL tasks in the mainstream research. Then all the mainstream state-of-the-art(SOTA) RL algorithms can be conveniently evaluated and compared. Therefore, our contributions mainly includes the following aspects: 1.single configured environment for all classification of SOTA RL algorithms; 2.combined environment of more than one classification RL algorithms; 3.the evaluation standard for all kinds of RL algorithms. With all these efforts, a possibility for breeding an AI with capability of general competency in a variety of tasks is provided, and maybe it will open up a new chapter for AI.

cs.LG↗

arXiv2021

Orientation-related twinning and dislocation glide in a Cantor High Entropy Alloy at room and cryogenic temperature studied by in situ TEM straining

In situ straining experiments were performed in a TEM on an equimolar CoCrFeMnNi (Cantor) high entropy alloy at room and cryogenic temperature. Perfect and partial dislocation activity were recorded in both cases. Twinning directly follows the development of partial dislocation shearing that has various origins (perfect dislocation splitting, anchoring). It is shown that, although twinning is more frequently observed at liquid nitrogen temperature, its prevalence depends mainly on crystal orientation. As a result, twinning and perfect dislocation plasticity are likely to occur jointly in random oriented polycrystals, even at early stages of deformation.

cond-mat.mtrl-sci↗

arXiv2021

4/-hybridization strength in CemMnIn3m+2n heavy-fermion compounds studied by Angle-Resolved Photoemission Spectroscopy

We systemically investigate the nature of Ce 4f electrons in structurally layered heavy-fermion compounds CcmMnIn3m+2n (with M =Co, Rh, Ir, and Pt, m=l, 2, n=0 - 2), at low temperature using on-resonance angle-resolved photoemission spectroscopy. Three heavy quasiparticle bands f^0, f^1_7/2 and f^1_5/2 are observed in all compounds, but their intensities and energy locations vary greatly with materials. The strong f^0 states imply that the localized electron behavior dominates the Ce 4f states. The Ce 4f electrons are partially hybridized with the conduction electrons, making them have the dual nature of localization and itinerant. Our quantitative comparison reveals that the f^1_5/2 / f^0 intensity ratio is more suitable to reflect the 4f-state hybridization strength.

cond-mat.str-el↗

arXiv2021

Cross-modality Discrepant Interaction Network for RGB-D Salient Object Detection

The popularity and promotion of depth maps have brought new vigor and vitality into salient object detection (SOD), and a mass of RGB-D SOD algorithms have been proposed, mainly concentrating on how to better integrate cross-modality features from RGB image and depth map. For the cross-modality interaction in feature encoder, existing methods either indiscriminately treat RGB and depth modalities, or only habitually utilize depth cues as auxiliary information of the RGB branch. Different from them, we reconsider the status of two modalities and propose a novel Cross-modality Discrepant Interaction Network (CDINet) for RGB-D SOD, which differentially models the dependence of two modalities according to the feature representations of different layers. To this end, two components are designed to implement the effective cross-modality interaction: 1) the RGB-induced Detail Enhancement (RDE) module leverages RGB modality to enhance the details of the depth features in low-level encoder stage. 2) the Depth-induced Semantic Enhancement (DSE) module transfers the object positioning and internal consistency of depth features to the RGB branch in high-level encoder stage. Furthermore, we also design a Dense Decoding Reconstruction (DDR) structure, which constructs a semantic block by combining multi-level encoder features to upgrade the skip connection in the feature decoding. Extensive experiments on five benchmark datasets demonstrate that our network outperforms $15$ state-of-the-art methods both quantitatively and qualitatively. Our code is publicly available at: https://rmcong.github.io/proj_CDINet.html.

cs.CV↗

arXiv2021

Geometrized effective-one-body formalism for extreme-mass-ratio limits: Generic orbits

Compact objects inspiraling into supermassive black holes, known as extreme-mass-ratio inspirals, are an important source for future space-borne gravitational-wave detectors. When constructing waveform templates, usually the adiabatic approximation is employed to treat the compact object as a test particle for a short duration, and the radiation reaction is reflected in the changes of the constants of motion. However, the mass of the compact object should have contributions to the background. In the present paper, employing the effective-one-body formalism, we analytically calculate the trajectories of a compact object around a massive Kerr black hole with generally three-dimensional orbits and express the fundamental orbital frequencies in explicit forms. In addition, by constructing an approximate "constant" similar to the Carter constant, we transfer the dynamical quantities such as energy, angular momentum, and the "Carter constant" to the semilatus rectum, eccentricity, and orbital inclination with mass-ratio corrections. The linear mass-ratio terms in the formalism may not be sufficient for accurate waveforms, but our analytical method for solving the equations of motion could be useful in various approaches to building waveform models.

gr-qc↗

arXiv2021

On Magic Distinct Labellings of Simple Graphs

A magic labelling of a graph $G$ with magic sum $s$ is a labelling of the edges of $G$ by nonnegative integers such that for each vertex $v\in V$, the sum of labels of all edges incident to $v$ is equal to the same number $s$. Stanley gave remarkable results on magic labellings, but the distinct labelling case is much more complicated. We consider the complete construction of all magic labellings of a given graph $G$. The idea is illustrated in detail by dealing with three regular graphs. We give combinatorial proofs. The structure result was used to enumerate the corresponding magic distinct labellings.

math.CO↗