Searcharxiv⌕ Search

arXiv subjects

Zhi Yang

Publications and source records attributed to Zhi Yang.

At least 109 records · Page 6Linked to original sources
arXiv2020

Photoproduction of hidden-bottom pentaquark and related topics

Due to the discovery of the hidden-charm pentaquark $P_c$ states by the LHCb collaboration, the interests on the candidates of hidden-bottom pentaquark $P_b$ states are increasing. They are anticipated to exist as the analogues of the $P_c$ states in the bottom sector and predicted by many models. We give an exploration of searching for a typical $P_b$ in the $γp \to Υp$ reaction, which shows a promising potential to observe it at an electron-ion collider. The possibility of searching for $P_b$ in open-bottom channels are also briefly discussed. Meanwhile, the $t$-channel non-resonant contribution, which in fact covers several interesting topics at low energies, is systematically investigated.

hep-ph↗
arXiv2020

Identify the hidden charm pentaquark signal from non-resonant background in electron-proton scattering

We study the electroproduction of the LHCb pentaquark states with the assumption that they are resonant states. The main concern here is to investigate the final state distribution in the phase space in order to extract the feeble pentaquark signal from the large non-resonant background. Our results show that the ratio of the signal to background would increase significantly with proper kinematic cut, which would be very helpful for future experimental analysis.

hep-ph↗
arXiv2019

MSD: Multi-Self-Distillation Learning via Multi-classifiers within Deep Neural Networks

As the development of neural networks, more and more deep neural networks are adopted in various tasks, such as image classification. However, as the huge computational overhead, these networks could not be applied on mobile devices or other low latency scenes. To address this dilemma, multi-classifier convolutional network is proposed to allow faster inference via early classifiers with the corresponding classifiers. These networks utilize sophisticated designing to increase the early classifier accuracy. However, naively training the multi-classifier network could hurt the performance (accuracy) of deep neural networks as early classifiers throughout interfere with the feature generation process. In this paper, we propose a general training framework named multi-self-distillation learning (MSD), which mining knowledge of different classifiers within the same network and increase every classifier accuracy. Our approach can be applied not only to multi-classifier networks, but also modern CNNs (e.g., ResNet Series) augmented with additional side branch classifiers. We use sampling-based branch augmentation technique to transform a single-classifier network into a multi-classifier network. This reduces the gap of capacity between different classifiers, and improves the effectiveness of applying MSD. Our experiments show that MSD improves the accuracy of various networks: enhancing the accuracy of every classifier significantly for existing multi-classifier network (MSDNet), improving vanilla single-classifier networks with internal classifiers with high accuracy, while also improving the final accuracy.

cs.CV↗
arXiv2019

Representation Learning of EHR Data via Graph-Based Medical Entity Embedding

Automatic representation learning of key entities in electronic health record (EHR) data is a critical step for healthcare informatics that turns heterogeneous medical records into structured and actionable information. Here we propose ME2Vec, an algorithmic framework for learning low-dimensional vectors of the most common entities in EHR: medical services, doctors, and patients. ME2Vec leverages diverse graph embedding techniques to cater for the unique characteristic of each medical entity. Using real-world clinical data, we demonstrate the efficacy of ME2Vec over competitive baselines on disease diagnosis prediction.

cs.LG↗
arXiv2019

Notes on the Causal Structure in a Tensor Network

In this paper we attempt to understand Lorentzian tensor networks, as a preparation for constructing tensor networks that can describe more exotic backgrounds such as black holes. To define notions of reference frames and switching of reference frames on a tensor network, we will borrow ideas from the algebraic quantum field theory literature. With these definitions, we construct simple examples of Lorentzian tensor networks and solve the spectrum for a choice of ``inertial frame'' based on Gaussian models of fermions and integrable models. In particular, the tensor network can be viewed as a periodically driven Floquet system, that by-pass the ``doubling problem'' and gives rise to fermions with exactly linear dispersion relations. We will find that a boost operator connecting different inertial frames, and notions of ``Rindler observers'' can be defined, and that important physics in Lorentz invariant QFT, such as the Unruh effect, can be captured by such skeleton of spacetime. We find interesting subtleties when the same approach is directly applied to bosons -- the operator algebra contains commutators that take the wrong sign -- resembling bosons behind horizons.

hep-th↗
arXiv2018

Embedding Push and Pull Search in the Framework of Differential Evolution for Solving Constrained Single-objective Optimization Problems

This paper proposes a push and pull search method in the framework of differential evolution (PPS-DE) to solve constrained single-objective optimization problems (CSOPs). More specifically, two sub-populations, including the top and bottom sub-populations, are collaborated with each other to search global optimal solutions efficiently. The top sub-population adopts the pull and pull search (PPS) mechanism to deal with constraints, while the bottom sub-population use the superiority of feasible solutions (SF) technique to deal with constraints. In the top sub-population, the search process is divided into two different stages --- push and pull stages.An adaptive DE variant with three trial vector generation strategies is employed in the proposed PPS-DE. In the top sub-population, all the three trial vector generation strategies are used to generate offsprings, just like in CoDE. In the bottom sub-population, a strategy adaptation, in which the trial vector generation strategies are periodically self-adapted by learning from their experiences in generating promising solutions in the top sub-population, is used to choose a suitable trial vector generation strategy to generate one offspring. Furthermore, a parameter adaptation strategy from LSHADE44 is employed in both sup-populations to generate scale factor $F$ and crossover rate $CR$ for each trial vector generation strategy. Twenty-eight CSOPs with 10-, 30-, and 50-dimensional decision variables provided in the CEC2018 competition on real parameter single objective optimization are optimized by the proposed PPS-DE. The experimental results demonstrate that the proposed PPS-DE has the best performance compared with the other seven state-of-the-art algorithms, including AGA-PPS, LSHADE44, LSHADE44+IDE, UDE, IUDE, $ε$MAg-ES and C$^2$oDE.

cs.NE↗
arXiv2018

Towards Efficient Large-Scale Graph Neural Network Computing

Recent deep learning models have moved beyond low-dimensional regular grids such as image, video, and speech, to high-dimensional graph-structured data, such as social networks, brain connections, and knowledge graphs. This evolution has led to large graph-based irregular and sparse models that go beyond what existing deep learning frameworks are designed for. Further, these models are not easily amenable to efficient, at scale, acceleration on parallel hardwares (e.g. GPUs). We introduce NGra, the first parallel processing framework for graph-based deep neural networks (GNNs). NGra presents a new SAGA-NN model for expressing deep neural networks as vertex programs with each layer in well-defined (Scatter, ApplyEdge, Gather, ApplyVertex) graph operation stages. This model not only allows GNNs to be expressed intuitively, but also facilitates the mapping to an efficient dataflow representation. NGra addresses the scalability challenge transparently through automatic graph partitioning and chunk-based stream processing out of GPU core or over multiple GPUs, which carefully considers data locality, data movement, and overlapping of parallel processing and data movement. NGra further achieves efficiency through highly optimized Scatter/Gather operators on GPUs despite its sparsity. Our evaluation shows that NGra scales to large real graphs that none of the existing frameworks can handle directly, while achieving up to about 4 times speedup even at small scales over the multiple-baseline design on TensorFlow.

cs.DC↗
arXiv2018

Deep Compressive Autoencoder for Action Potential Compression in Large-Scale Neural Recording

Understanding the coordinated activity underlying brain computations requires large-scale, simultaneous recordings from distributed neuronal structures at a cellular-level resolution. One major hurdle to design high-bandwidth, high-precision, large-scale neural interfaces lies in the formidable data streams that are generated by the recorder chip and need to be online transferred to a remote computer. The data rates can require hundreds to thousands of I/O pads on the recorder chip and power consumption on the order of Watts for data streaming alone. We developed a deep learning-based compression model to reduce the data rate of multichannel action potentials. The proposed model is built upon a deep compressive autoencoder (CAE) with discrete latent embeddings. The encoder is equipped with residual transformations to extract representative features from spikes, which are mapped into the latent embedding space and updated via vector quantization (VQ). The decoder network reconstructs spike waveforms from the quantized latent embeddings. Experimental results show that the proposed model consistently outperforms conventional methods by achieving much higher compression ratios (20-500x) and better or comparable reconstruction accuracies. Testing results also indicate that CAE is robust against a diverse range of imperfections, such as waveform variation and spike misalignment, and has minor influence on spike sorting accuracy. Furthermore, we have estimated the hardware cost and real-time performance of CAE and shown that it could support thousands of recording channels simultaneously without excessive power/heat dissipation. The proposed model can reduce the required data transmission bandwidth in large-scale recording experiments and maintain good signal qualities. The code of this work has been made available at https://github.com/tong-wu-umn/spike-compression-autoencoder

cs.NE↗
arXiv2018

Blockage Modeling for Inter-layer UAVs Communications in Urban Environments

The impact of buildings blockage on the UAVs communications in different UAVs heights is studied, when UAVs fly in a given rectangular urban area. This paper analyzes the properties of blockage behavior of communication links between UAVs from layers of different heights, including two main stochastic properties, namely the blockage (or LOS) probabilities, and the state transition models (birth-death process of LOS propagation). Based on stochastic geometry methods, the relation between LOS probability and the parameters of the region and the buildings is derived and verified with simulations. Furthermore, based on the simulations of the channel state transition process, we also find the birth-death process can be modeled as a distance continuous Markov process, and the transition rates are also extracted with their relation with the height of layers.

cs.IT↗
arXiv2018

Channel Model in Urban Environment for Unmanned Aerial Vehicle Communications

In order to develop and analyze reliable communications links for unmanned aerial vehicles (UAVs), accurate models for the propagation channel are required. The radio channel properties in the urban scenario are different from those in the suburb scenario and open area due to so many scattering paths from office buildings, especially when the UAV flies in the low altitude. We took some measurement campaigns on the campus of Tsinghua University with crowded apartments and office buildings. Based on the measurement result we extract the main parameters of pathloss model, and propose a simplified Saleh-Valenzuela (SV) model with specific parameters. The typical scene of central lawn is compared with the office buildings in the analysis of K-factor and root-mean-square (RMS) delay spread.

cs.IT↗
arXiv2018

Coverage Probability Analysis of UAV Cellular Networks in Urban Environments

In this paper, we study coverage probabilities of the UAV-assisted cellular network modeled by 2-dimension (2D) Poisson point process. The cellular user is assumed to connect to the nearest aerial base station. We derive the explicit expressions for the downlink coverage probability for the Rayleigh fading channel. Furthermore, we explore the behavior of performance when taking the property of air-to-ground channel into consideration. Our analytical and numerical results show that the coverage probability is affected by UAV height, pathloss exponent and UAV density. To maximize the coverage probability, the optimal height and density of UAVs are studied, which could be beneficial for the UAV deployment design.

cs.IT↗
arXiv2018

Achieving Super-Resolution with Redundant Sensing

Analog-to-digital (quantization) and digital-to-analog (de-quantization) conversion are fundamental operations of many information processing systems. In practice, the precision of these operations is always bounded, first by the random mismatch error (ME) occurred during system implementation, and subsequently by the intrinsic quantization error (QE) determined by the system architecture itself. In this manuscript, we present a new mathematical interpretation of the previously proposed redundant sensing (RS) architecture that not only suppresses ME but also allows achieving an effective resolution exceeding the system's intrinsic resolution, i.e. super-resolution (SR). SR is enabled by an endogenous property of redundant structures regarded as "code diffusion" where the references' value spreads into the neighbor sample space as a result of ME. The proposed concept opens the possibility for a wide range of applications in low-power fully-integrated sensors and devices where the cost-accuracy trade-off is inevitable.

cs.IT↗
arXiv2018

A Game Theoretic Model for the Formation of Navigable Small-World Networks --- the Tradeoff between Distance and Reciprocity

Kleinberg proposed a family of small-world networks to explain the navigability of large-scale real-world social networks. However, the underlying mechanism that drives real networks to be navigable is not yet well understood. In this paper, we present a game theoretic model for the formation of navigable small world networks. We model the network formation as a Distance-Reciprocity Balanced (DRB) game in which people seek for both high reciprocity and long-distance relationships. We show that the game has only two Nash equilibria: One is the navigable small-world network, and the other is the random network in which each node connects with each other node with equal probability. We further show that the navigable small world is very stable --- (a) no collusion of any size would benefit from deviating from it; and (b) after an arbitrary deviations of a large random set of nodes, the network would return to the navigable small world as soon as every node takes one best-response step. In contrast, for the random network, a small group collusion or random perturbations is guaranteed to move the network to the navigable network as soon as every node takes one best-response step. Moreover, we show that navigable small world has much better social welfare than the random network, and provide the price-of-anarchy and price-of-stability results of the game. Our empirical evaluation demonstrates that the system always converges to the navigable network even when limited or no information about other players' strategies is available, and the DRB game simulated on the real-world network leads to navigability characteristic that is very close to that of the real network. Our theoretical and empirical analyses provide important new insight on the connection between distance, reciprocity and navigability in social networks.

cs.SI↗
arXiv2018

Gauge choices and Entanglement Entropy of two dimensional lattice gauge fields

In this paper, we explore the question of how different gauge choices in a gauge theory affect the tensor product structure of the Hilbert space in configuration space. In particular, we study the Coulomb gauge and observe that the naive gauge potential degrees of freedom cease to be local operators as soon as we impose the Dirac brackets. We construct new local set of operators and compute the entanglement entropy according to this algebra in $2+1$ dimensions. We find that our proposal would lead to an entanglement entropy that behave very similar to a single scalar degree of freedom if we do not include further centers, but approaches that of a gauge field if we include non-trivial centers. We explore also the situation where the gauge field is Higgsed, and construct a local operator algebra that again requires some deformation. This should give us some insight into interpreting the entanglement entropy in generic gauge theories and perhaps also in gravitational theories.

hep-th↗
arXiv2018

A Bio-inspired Redundant Sensing Architecture

Sensing is the process of deriving signals from the environment that allows artificial systems to interact with the physical world. The Shannon theorem specifies the maximum rate at which information can be acquired. However, this upper bound is hard to achieve in many man-made systems. The biological visual systems, on the other hand, have highly efficient signal representation and processing mechanisms that allow precise sensing. In this work, we argue that redundancy is one of the critical characteristics for such superior performance. We show architectural advantages by utilizing redundant sensing, including correction of mismatch error and significant precision enhancement. For a proof-of-concept demonstration, we have designed a heuristic-based analog-to-digital converter - a zero-dimensional quantizer. Through Monte Carlo simulation with the error probabilistic distribution as a priori, the performance approaching the Shannon limit is feasible. In actual measurements without knowing the error distribution, we observe at least 2-bit extra precision. The results may also help explain biological processes including the dominance of binocular vision, the functional roles of the fixational eye movements, and the structural mechanisms allowing hyperacuity.

cs.NE↗
arXiv2018

Advancing System Performance with Redundancy: From Biological to Artificial Designs

Redundancy is a fundamental characteristic of many biological processes such as those in the genetic, visual, muscular and nervous system; yet its function has not been fully understood. The conventional interpretation of redundancy is that it serves as a fault-tolerance mechanism, which leads to redundancy's de facto application in man-made systems for reliability enhancement. On the contrary, our previous works have demonstrated an example where redundancy can be engineered solely for enhancing other aspects of the system, namely accuracy and precision. This design was inspired by the binocular structure of the human vision which we believe may share a similar operation. In this paper, we present a unified theory describing how such utilization of redundancy is feasible through two complementary mechanisms: representational redundancy (RPR) and entangled redundancy (ETR). Besides the previous works, we point out two additional examples where our new understanding of redundancy can be applied to justify a system's superior performance. One is the human musculoskeletal system (HMS) - a biological instance, and one is the deep residual neural network (ResNet) - an artificial counterpart. We envision that our theory would provide a framework for the future development of bio-inspired redundant artificial systems as well as assist the studies of the fundamental mechanisms governing various biological processes.

cs.IT↗
arXiv2017

Isospin analysis of $B\to D^*\bar{D}K$ and the absence of the $Z_c(3900)$ in $B$ decays

We study the isospin amplitudes in the exclusive $B$ to $D^*\bar{D}K$ decay process and fit the available $D^{*0}\bar{D}^0$ invariant mass distributions near threshold. The analysis demonstrates that the production of the isospin triplet $D^*\bar{D}$ state is highly suppressed compared to the isospin singlet one. That explains why the $Z_c(3900)$ has not been found in $B$ decays. In addition, the production of the negative charge-parity state might be further suppressed in the heavy quark limit. These two reasons which are based on the molecular assumption offer the first explanation why the $Z_c(3900)$ is absent in $B$ decays. Further studies of the absence from both the experimental and the theoretical side is extremely important for understanding the nature of the $X(3872)$ and the $Z_c(3900)$.

hep-ph↗