arXiv subjects

Chi Zhang

Publications and source records attributed to Chi Zhang.

At least 721 records · Page 40Linked to original sources

arXiv2021

On finite groups factorized by $σ$-nilpotent subgroups

Let $G$ be a finite group and $σ=\{σ_{i}|i\in I\}$ be a partition of the set of all primes $\mathbb{P}$, that is, $\mathbb{P}=\bigcup_{i\in I}σ_{i}$ and $σ_{i}\cap σ_{j}=\emptyset$ for all $i\neq j$. A chief factor $H/K$ of $G$ is said to be $σ$-central in $G$, if the semidirect product $(H/K)\rtimes(G/C_G(H/K))$ is a $σ_i$-group for some $i\in I$. The group $G$ is said to be $σ$-nilpotent if either $G=1$ or every chief factor of $G$ is $σ$-central. In this paper, we study the properties of a finite group $G=AB$, factorized by two $σ$-nilpotent subgroups $A$ and $B$, and also generalize some known results.

math.GR↗

arXiv2021

Quasi-two-dimensional heterostructures (K$M_{1-x}$Te)(LaTe$_{3}$) ($M$ = Mn, Zn) with charge density waves

Layered heterostructure materials with two different functional building blocks can teach us about emergent physical properties and phenomena arising from interactions between the layers. We report the intergrowth compounds KLa$M$$_{1-x}$Te$_{4}$ ($M$ = Mn, Zn; $x\approx$ 0.35) featuring two chemically distinct alternating layers [LaTe$_3$] and [K$M$$_{1-x}$Te]. Their crystal structures are incommensurate, determined by single X-ray diffraction for the Mn compound and transmission electron microscope (TEM) study for the Zn compound. KLaMn$_{1-x}$Te$_{4}$ crystallizes in the orthorhombic superspace group $Pmnm$(01/2$γ$)$s$00 with lattice parameters $a$ = 4.4815(3) Å, $b$ = 21.6649(16) Å and $c$ = 4.5220(3) Å. It exhibits charge density wave (CDW) order at room temperature with a modulation wave vector $\mathbf{q}$ = 1/2$\mathbf{b}$* + 0.3478$\mathbf{c}$* originating from electronic instability of Te-square nets in [LaTe$_{3}$] layers. The Mn analog exhibits a cluster spin glass behavior with spin freezing temperature $T_{\mathrm{f}}$ $\approx$ 5 K attributed to disordered Mn vacancies and competing magnetic interactions in the [Mn$_{1-x}$Te] layers. The Zn analog also has charge density wave order at room temperature with a similar $\mathbf{q}$-vector having the $\mathbf{c}$* component ~ 0.346 confirmed by selected-area electron diffraction (SAED). Electron transfer from [K$M_{1-x}$Te] to [LaTe$_{3}$] layers exists in KLa$M_{1-x}$Te$_{4}$, leading to an enhanced electronic specific heat coefficient. The resistivities of KLa$M_{1-x}$Te$_{4}$ ($M$ = Mn, Zn) exhibit metallic behavior at high temperatures and an upturn at low temperatures, suggesting partial localization of carriers in the [LaTe$_{3}$] layers with some degree of disorder associated with the $M$ atom vacancies in the [$M_{1-x}$Te] layers.

cond-mat.mtrl-sci↗

arXiv2021

Few-Shot Incremental Learning with Continually Evolved Classifiers

Few-shot class-incremental learning (FSCIL) aims to design machine learning algorithms that can continually learn new concepts from a few data points, without forgetting knowledge of old classes. The difficulty lies in that limited data from new classes not only lead to significant overfitting issues but also exacerbate the notorious catastrophic forgetting problems. Moreover, as training data come in sequence in FSCIL, the learned classifier can only provide discriminative information in individual sessions, while FSCIL requires all classes to be involved for evaluation. In this paper, we address the FSCIL problem from two aspects. First, we adopt a simple but effective decoupled learning strategy of representations and classifiers that only the classifiers are updated in each incremental session, which avoids knowledge forgetting in the representations. By doing so, we demonstrate that a pre-trained backbone plus a non-parametric class mean classifier can beat state-of-the-art methods. Second, to make the classifiers learned on individual sessions applicable to all classes, we propose a Continually Evolved Classifier (CEC) that employs a graph model to propagate context information between classifiers for adaptation. To enable the learning of CEC, we design a pseudo incremental learning paradigm that episodically constructs a pseudo incremental learning task to optimize the graph parameters by sampling data from the base dataset. Experiments on three popular benchmark datasets, including CIFAR100, miniImageNet, and Caltech-USCD Birds-200-2011 (CUB200), show that our method significantly outperforms the baselines and sets new state-of-the-art results with remarkable advantages.

cs.CV↗

arXiv2021

Efficient DETR: Improving End-to-End Object Detector with Dense Prior

The recently proposed end-to-end transformer detectors, such as DETR and Deformable DETR, have a cascade structure of stacking 6 decoder layers to update object queries iteratively, without which their performance degrades seriously. In this paper, we investigate that the random initialization of object containers, which include object queries and reference points, is mainly responsible for the requirement of multiple iterations. Based on our findings, we propose Efficient DETR, a simple and efficient pipeline for end-to-end object detection. By taking advantage of both dense detection and sparse set detection, Efficient DETR leverages dense prior to initialize the object containers and brings the gap of the 1-decoder structure and 6-decoder structure. Experiments conducted on MS COCO show that our method, with only 3 encoder layers and 1 decoder layer, achieves competitive performance with state-of-the-art object detection methods. Efficient DETR is also robust in crowded scenes. It outperforms modern detectors on CrowdHuman dataset by a large margin.

cs.CV↗

arXiv2021

Parameterization of nearshore wave breaker index

The performances of phase-averaged parametric nearshore wave transformation models depend significantly on a reliable estimate of the wave breaker index r (the breaker height-to-depth ratio), a free model parameter that essentially controls the amount of breaking energy dissipation. While the previous studies have suggested separate relationships between r and the offshore wave steepness (s0) or the normalized local water depth (kh), the parameterization of r still requires further investigation considering a wider variety of conditions and a sounder physical basis. In this study, we use the field datasets of wave height and the inverse modelling approach to reveal a composite dependence of r on both s0 and kh. Specifically, the results show a positive dependence of r on kh for larger s0, and a negative dependence of r on kh for smaller s0. Based on such composite relationships, a new r formula is proposed, and its performance is verified against the available datasets of wave height in three coasts and 14 laboratory tests. Implementation of this new formula in a parametric wave model leads to the error reduction of wave height prediction by 10~24% (mean = 19%) relative to seven widely used models in literatures. In particular, a remarkably higher model accuracy is obtained under wave conditions with small offshore wave steepness, which is important for studying onshore sediment transport and beach recovery. Two counteractive physical mechanisms for wave nonlinearity effects, namely the breaking intensification mechanism and the breaking resistance mechanism, are suggested to explain the opposite r-kh relationships within different ranges of s0.

physics.ao-ph↗

arXiv2021

Nonreciprocal Multi-mode and Indirect Couplings in Cavity Magnonics

We investigate the magnon-photon couplings by employing a small magnet within an irregular resonant cavity, which leads to a desirable nonreciprocity with a big isolation ratio. Moreover, the higher-order couplings between the spin wave modes with the polarized photon modes also exhibit the nonreciprocity. These couplings between polarized photon and spin waves could be regarded as an indirect multi-modes coupling between the ferromagnetic resonance (FMR) mode and spin wave mode magnons mediated by the cavity mode photons. We also derive a coupling matrix to predict the characteristics of this kind of indirect coupling. The existence of the indirect couplings broaden the field range of the nonreciprocity of the system. The achieved nonreciprocal multi-mode magnon-photon couplings in a single system offer a feasible method to improve the signal transmission quality.

physics.app-ph↗

arXiv2021

Congestion-aware Multi-agent Trajectory Prediction for Collision Avoidance

Predicting agents' future trajectories plays a crucial role in modern AI systems, yet it is challenging due to intricate interactions exhibited in multi-agent systems, especially when it comes to collision avoidance. To address this challenge, we propose to learn congestion patterns as contextual cues explicitly and devise a novel "Sense--Learn--Reason--Predict" framework by exploiting advantages of three different doctrines of thought, which yields the following desirable benefits: (i) Representing congestion as contextual cues via latent factors subsumes the concept of social force commonly used in physics-based approaches and implicitly encodes the distance as a cost, similar to the way a planning-based method models the environment. (ii) By decomposing the learning phases into two stages, a "student" can learn contextual cues from a "teacher" while generating collision-free trajectories. To make the framework computationally tractable, we formulate it as an optimization problem and derive an upper bound by leveraging the variational parametrization. In experiments, we demonstrate that the proposed model is able to generate collision-free trajectory predictions in a synthetic dataset designed for collision avoidance evaluation and remains competitive on the commonly used NGSIM US-101 highway dataset.

cs.RO↗

arXiv2021

ACRE: Abstract Causal REasoning Beyond Covariation

Causal induction, i.e., identifying unobservable mechanisms that lead to the observable relations among variables, has played a pivotal role in modern scientific discovery, especially in scenarios with only sparse and limited data. Humans, even young toddlers, can induce causal relationships surprisingly well in various settings despite its notorious difficulty. However, in contrast to the commonplace trait of human cognition is the lack of a diagnostic benchmark to measure causal induction for modern Artificial Intelligence (AI) systems. Therefore, in this work, we introduce the Abstract Causal REasoning (ACRE) dataset for systematic evaluation of current vision systems in causal induction. Motivated by the stream of research on causal discovery in Blicket experiments, we query a visual reasoning system with the following four types of questions in either an independent scenario or an interventional scenario: direct, indirect, screening-off, and backward-blocking, intentionally going beyond the simple strategy of inducing causal relationships by covariation. By analyzing visual reasoning architectures on this testbed, we notice that pure neural models tend towards an associative strategy under their chance-level performance, whereas neuro-symbolic combinations struggle in backward-blocking reasoning. These deficiencies call for future research in models with a more comprehensive capability of causal induction.

cs.CV↗

arXiv2021

Dynamic Metric Learning: Towards a Scalable Metric Space to Accommodate Multiple Semantic Scales

This paper introduces a new fundamental characteristic, \ie, the dynamic range, from real-world metric tools to deep visual recognition. In metrology, the dynamic range is a basic quality of a metric tool, indicating its flexibility to accommodate various scales. Larger dynamic range offers higher flexibility. In visual recognition, the multiple scale problem also exist. Different visual concepts may have different semantic scales. For example, ``Animal'' and ``Plants'' have a large semantic scale while ``Elk'' has a much smaller one. Under a small semantic scale, two different elks may look quite \emph{different} to each other . However, under a large semantic scale (\eg, animals and plants), these two elks should be measured as being \emph{similar}. %We argue that such flexibility is also important for deep metric learning, because different visual concepts indeed correspond to different semantic scales. Introducing the dynamic range to deep metric learning, we get a novel computer vision task, \ie, the Dynamic Metric Learning. It aims to learn a scalable metric space to accommodate visual concepts across multiple semantic scales. Based on three types of images, \emph{i.e.}, vehicle, animal and online products, we construct three datasets for Dynamic Metric Learning. We benchmark these datasets with popular deep metric learning methods and find Dynamic Metric Learning to be very challenging. The major difficulty lies in a conflict between different scales: the discriminative ability under a small scale usually compromises the discriminative ability under a large one, and vice versa. As a minor contribution, we propose Cross-Scale Learning (CSL) to alleviate such conflict. We show that CSL consistently improves the baseline on all the three datasets. The datasets and the code will be publicly available at https://github.com/SupetZYK/DynamicMetricLearning.

cs.CV↗

arXiv2021

Building HVAC Scheduling Using Reinforcement Learning via Neural Network Based Model Approximation

Buildings sector is one of the major consumers of energy in the United States. The buildings HVAC (Heating, Ventilation, and Air Conditioning) systems, whose functionality is to maintain thermal comfort and indoor air quality (IAQ), account for almost half of the energy consumed by the buildings. Thus, intelligent scheduling of the building HVAC system has the potential for tremendous energy and cost savings while ensuring that the control objectives (thermal comfort, air quality) are satisfied. Recently, several works have focused on model-free deep reinforcement learning based techniques such as Deep Q-Network (DQN). Such methods require extensive interactions with the environment. Thus, they are impractical to implement in real systems due to low sample efficiency. Safety-aware exploration is another challenge in real systems since certain actions at particular states may result in catastrophic outcomes. To address these issues and challenges, we propose a model-based reinforcement learning approach that learns the system dynamics using a neural network. Then, we adopt Model Predictive Control (MPC) using the learned system dynamics to perform control with random-sampling shooting method. To ensure safe exploration, we limit the actions within safe range and the maximum absolute change of actions according to prior knowledge. We evaluate our ideas through simulation using widely adopted EnergyPlus tool on a case study consisting of a two zone data-center. Experiments show that the average deviation of the trajectories sampled from the learned dynamics and the ground truth is below $20\%$. Compared with baseline approaches, we reduce the total energy consumption by $17.1\% \sim 21.8\%$. Compared with model-free reinforcement learning approach, we reduce the required number of training steps to converge by 10x.

eess.SY↗

arXiv2021

Matched Queues with Matching Batch Pair (m, n)

In this paper, we discuss an interesting but challenging bilateral stochastically matching problem: A more general matched queue with matching batch pair (m, n) and two types (i.e., types A and B) of impatient customers, where the arrivals of A- and B-customers are both Poisson processes, m A-customers and n B-customers are matched as a group which leaves the system immediately, and the customers' impatient behavior is to guarantee the stability of the system. We show that this matched queue can be expressed as a novel bidirectional level-dependent quasi-birth-and-death (QBD) process. Based on this, we provide a detailed analysis for this matched queue, including the system stability, the average stationary queue lengthes, and the average sojourn time of any A-customer or B-customer. We believe that the methodology and results developed in this paper can be applicable to dealing with more general matched queueing systems, which are widely encountered in various practical areas, for example, sharing economy, ridesharing platform, bilateral market, organ transplantation, taxi services, assembly systems, and so on.

math.PR↗

arXiv2021

A splitting random-choice dynamic relaxation method for smoothed particle hydrodynamics

For conventional smoothed particle hydrodynamics (SPH), obtaining the static solution of a problem is time-consuming. To address this drawback, we propose an efficient dynamic relaxation method by adding large artificial-viscosity-based damping into the momentum conservation equation. Then, operator splitting methods are introduced to discretize the added viscous term for relaxing the time-step limit. To further improve the convergence rate, a random-choice strategy is adopted, in which the viscous term is imposed randomly rather than at every time step. In addition, to avoid the thread-conflict induced by applying shared-memory parallelization to accelerate implicit method, a splitting cell-linked list scheme is devised. A number of benchmark tests suggest that the present method helps systems achieve equilibrium state efficiently.

cs.CE↗

arXiv2021

FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding

Emerging interests have been brought to recognize previously unseen objects given very few training examples, known as few-shot object detection (FSOD). Recent researches demonstrate that good feature embedding is the key to reach favorable few-shot learning performance. We observe object proposals with different Intersection-of-Union (IoU) scores are analogous to the intra-image augmentation used in contrastive approaches. And we exploit this analogy and incorporate supervised contrastive learning to achieve more robust objects representations in FSOD. We present Few-Shot object detection via Contrastive proposals Encoding (FSCE), a simple yet effective approach to learning contrastive-aware object proposal encodings that facilitate the classification of detected objects. We notice the degradation of average precision (AP) for rare objects mainly comes from misclassifying novel instances as confusable classes. And we ease the misclassification issues by promoting instance level intra-class compactness and inter-class variance via our contrastive proposal encoding loss (CPE loss). Our design outperforms current state-of-the-art works in any shot and all data splits, with up to +8.8% on standard benchmark PASCAL VOC and +2.7% on challenging COCO benchmark. Code is available at: https: //github.com/MegviiDetection/FSCE

cs.CV↗

arXiv2021

Experimental Side-Channel-Free Quantum Key Distribution

Quantum key distribution can provide unconditionally secure key exchange for remote users in theory. In practice, however, in most quantum key distribution systems, quantum hackers might steal the secure keys by listening to the side channels in the source, such as the photon frequency spectrum, emission time, propagation direction, spatial angular momentum, and so on. It is hard to prevent such kinds of attacks because side channels may exist in any of the encoding space whether the designers take care of or not. Here we report an experimental realization of a side-channel-free quantum key distribution protocol which is not only measurement-device-independent, but also immune to all side-channel attacks in the source. We achieve a secure key rate of 4.80e-7 per pulse through 50 km fiber spools.

quant-ph↗

arXiv2021

End-to-End Human Object Interaction Detection with HOI Transformer

We propose HOI Transformer to tackle human object interaction (HOI) detection in an end-to-end manner. Current approaches either decouple HOI task into separated stages of object detection and interaction classification or introduce surrogate interaction problem. In contrast, our method, named HOI Transformer, streamlines the HOI pipeline by eliminating the need for many hand-designed components. HOI Transformer reasons about the relations of objects and humans from global image context and directly predicts HOI instances in parallel. A quintuple matching loss is introduced to force HOI predictions in a unified way. Our method is conceptually much simpler and demonstrates improved accuracy. Without bells and whistles, HOI Transformer achieves $26.61\% $ $ AP $ on HICO-DET and $52.9\%$ $AP_{role}$ on V-COCO, surpassing previous methods with the advantage of being much simpler. We hope our approach will serve as a simple and effective alternative for HOI tasks. Code is available at https://github.com/bbepoch/HoiTransformer .

cs.CV↗

arXiv2021

On Instabilities of Conventional Multi-Coil MRI Reconstruction to Small Adverserial Perturbations

Although deep learning (DL) has received much attention in accelerated MRI, recent studies suggest small perturbations may lead to instabilities in DL-based reconstructions, leading to concern for their clinical application. However, these works focus on single-coil acquisitions, which is not practical. We investigate instabilities caused by small adversarial attacks for multi-coil acquisitions. Our results suggest that, parallel imaging and multi-coil CS exhibit considerable instabilities against small adversarial perturbations.

eess.IV↗

arXiv2021

The symbol and alphabet of two-loop NMHV amplitudes from $\bar{Q}$ equations

We study the symbol and the alphabet for two-loop NMHV amplitudes in planar ${\cal N}=4$ super-Yang-Mills from the $\bar{Q}$ equations, which provide a first-principle method for computing multi-loop amplitudes. Starting from one-loop N${}^2$MHV ratio functions, we explain in detail how to use $\bar{Q}$ equations to obtain the total differential of two-loop $n$-point NMHV amplitudes, whose symbol contains letters that are algebraic functions of kinematics for $n\geq 8$. We present explicit formula with nice patterns for the part of the symbol involving algebraic letters for all multiplicities, and we find $17-2m$ multiplicative-independent letters for a given square root of Gram determinant, with $0\leq m\leq 4$ depending on the number of particles involved in the square root. We also observe that these algebraic letters can be found as poles of one-loop four-mass leading singularities with MHV or NMHV trees. As a byproduct of our algebraic results, we find a large class of components of two-loop NMHV, which can be written as differences of two double-pentagon integrals, particularly simple and absent of square roots. As an example, we present the complete symbol for $n=9$ whose alphabet contains $59\times 9$ rational letters, in addition to the $11 \times 9$ independent algebraic ones. We also give all-loop NMHV last-entry conditions for all multiplicities.

hep-th↗

arXiv2021

Conditional Gaussian Distribution Learning for Open Set Recognition

Deep neural networks have achieved state-of-the-art performance in a wide range of recognition/classification tasks. However, when applying deep learning to real-world applications, there are still multiple challenges. A typical challenge is that unknown samples may be fed into the system during the testing phase and traditional deep neural networks will wrongly recognize the unknown sample as one of the known classes. Open set recognition is a potential solution to overcome this problem, where the open set classifier should have the ability to reject unknown samples as well as maintain high classification accuracy on known classes. The variational auto-encoder (VAE) is a popular model to detect unknowns, but it cannot provide discriminative representations for known classification. In this paper, we propose a novel method, Conditional Gaussian Distribution Learning (CGDL), for open set recognition. In addition to detecting unknown samples, this method can also classify known samples by forcing different latent features to approximate different Gaussian models. Meanwhile, to avoid information hidden in the input vanishing in the middle layers, we also adopt the probabilistic ladder architecture to extract high-level abstract features. Experiments on several standard image datasets reveal that the proposed method significantly outperforms the baseline method and achieves new state-of-the-art results.

cs.LG↗