arXiv subjects

Yang Li

Publications and source records attributed to Yang Li.

At least 1,459 records · Page 81Linked to original sources

arXiv2018

Model Confidence Bounds for Variable Selection

In this article, we introduce the concept of model confidence bounds (MCB) for variable selection in the context of nested models. Similarly to the endpoints in the familiar confidence interval for parameter estimation, the MCB identifies two nested models (upper and lower confidence bound models) containing the true model at a given level of confidence. Instead of trusting a single selected model obtained from a given model selection method, the MCB proposes a group of nested models as candidates and the MCB's width and composition enable the practitioner to assess the overall model selection uncertainty. A new graphical tool --- the model uncertainty curve (MUC) --- is introduced to visualize the variability of model selection and to compare different model selection procedures. The MCB methodology is implemented by a fast bootstrap algorithm that is shown to yield the correct asymptotic coverage under rather general conditions. Our Monte Carlo simulations and real data examples confirm the validity and illustrate the advantages of the proposed method.

stat.ME↗

arXiv2018

When Do Birds of a Feather Flock Together? k-Means, Proximity, and Conic Programming

Given a set of data, one central goal is to group them into clusters based on some notion of similarity between the individual objects. One of the most popular and widely-used approaches is k-means despite the computational hardness to find its global minimum. We study and compare the properties of different convex relaxations by relating them to corresponding proximity conditions, an idea originally introduced by Kumar and Kannan. Using conic duality theory, we present an improved proximity condition under which the Peng-Wei relaxation of k-means recovers the underlying clusters exactly. Our proximity condition improves upon Kumar and Kannan, and is comparable to that of Awashti and Sheffet where proximity conditions are established for projective k-means. In addition, we provide a necessary proximity condition for the exactness of the Peng-Wei relaxation. For the special case of equal cluster sizes, we establish a different and completely localized proximity condition under which the Amini-Levina relaxation yields exact clustering, thereby having addressed an open problem by Awasthi and Sheffet in the balanced case. Our framework is not only deterministic and model-free but also comes with a clear geometric meaning which allows for further analysis and generalization. Moreover, it can be conveniently applied to analyzing various data generative models such as the stochastic ball models and Gaussian mixture models. With this method, we improve the current minimum separation bound for the stochastic ball models and achieve the state-of-the-art results of learning Gaussian mixture models.

math.OC↗

arXiv2018

Time-Dependent Representation for Neural Event Sequence Prediction

Existing sequence prediction methods are mostly concerned with time-independent sequences, in which the actual time span between events is irrelevant and the distance between events is simply the difference between their order positions in the sequence. While this time-independent view of sequences is applicable for data such as natural languages, e.g., dealing with words in a sentence, it is inappropriate and inefficient for many real world events that are observed and collected at unequally spaced points of time as they naturally arise, e.g., when a person goes to a grocery store or makes a phone call. The time span between events can carry important information about the sequence dependence of human behaviors. In this work, we propose a set of methods for using time in sequence prediction. Because neural sequence models such as RNN are more amenable for handling token-like input, we propose two methods for time-dependent event representation, based on the intuition on how time is tokenized in everyday life and previous work on embedding contextualization. We also introduce two methods for using next event duration as regularization for training a sequence prediction model. We discuss these methods based on recurrent neural nets. We evaluate these methods as well as baseline models on five datasets that resemble a variety of sequence prediction tasks. The experiments revealed that the proposed methods offer accuracy gain over baseline models in a range of settings.

cs.LG↗

arXiv2018

A readout method based on 10 Gigabit Ethernet for Sipixel detector

With the rapid development of network protocol, TCP/IP has been widely applied in various OS systems because of its advantages of high-speed network transmission and standardization. Furthermore, TCP/IP has the functions of retransmission and flow control, so it can ensure data's security and stability. In recent years, as a transmission Field-bus, it is also widely used in high-energy physics detectors. This article describes a readout method using 10 gigabit network processor applied in silicon pixel detector which is a pixel detector designed for the High Energy Photon Source (HEPS) in China. The PHY layer and the MAC layer of this hardware are both instantiated in FPGA. The logic of hardware can process and read out the data of the detectors. Meanwhile, it wraps up data and send them to TOE (TCP Offload Engine). A kind of bus control protocol interface is also provided, which can facilitate the host computer to read and write the specific register directly through the UDP protocol. Only a piece of FPGA can complete the processing of the TCP and UDP packets. It is important for the miniaturization of the detectors.

cs.NI↗

arXiv2018

GroupReduce: Block-Wise Low-Rank Approximation for Neural Language Model Shrinking

Model compression is essential for serving large deep neural nets on devices with limited resources or applications that require real-time responses. As a case study, a state-of-the-art neural language model usually consists of one or more recurrent layers sandwiched between an embedding layer used for representing input tokens and a softmax layer for generating output tokens. For problems with a very large vocabulary size, the embedding and the softmax matrices can account for more than half of the model size. For instance, the bigLSTM model achieves state-of- the-art performance on the One-Billion-Word (OBW) dataset with around 800k vocabulary, and its word embedding and softmax matrices use more than 6GBytes space, and are responsible for over 90% of the model parameters. In this paper, we propose GroupReduce, a novel compression method for neural language models, based on vocabulary-partition (block) based low-rank matrix approximation and the inherent frequency distribution of tokens (the power-law distribution of words). The experimental results show our method can significantly outperform traditional compression methods such as low-rank approximation and pruning. On the OBW dataset, our method achieved 6.6 times compression rate for the embedding and softmax matrices, and when combined with quantization, our method can achieve 26 times compression rate, which translates to a factor of 12.8 times compression for the entire model with very little degradation in perplexity.

cs.CL↗

arXiv2018

Techniques for Efficiently Handling Power Surges in Fuel Cell Powered Data Centers: Modeling, Analysis, Results

Fuel cells are a promising power source for future data centers, offering high energy efficiency, low greenhouse gas emissions, and high reliability. However, due to mechanical limitations related to fuel delivery, fuel cells are slow to adjust to sudden increases in data center power demands, which can result in temporary power shortfalls. To mitigate the impact of power shortfalls, prior work has proposed to either perform power capping by throttling the servers, or to leverage energy storage devices (ESDs) that can temporarily provide enough power to make up for the shortfall while the fuel cells ramp up power generation. Both approaches have disadvantages: power capping conservatively limits server performance and can lead to service level agreement (SLA) violations, while ESD-only solutions must significantly overprovision the energy storage device capacity to tolerate the shortfalls caused by the worst-case (i.e., largest) power surges, which greatly increases the total cost of ownership (TCO). We propose SizeCap, the first ESD sizing framework for fuel cell powered data centers, which coordinates ESD sizing with power capping to enable a cost-effective solution to power shortfalls in data centers. SizeCap sizes the ESD just large enough to cover the majority of power surges, but not the worst-case surges that occur infrequently, to greatly reduce TCO. It then uses the smaller capacity ESD in conjunction with power capping to cover the power shortfalls caused by the worst-case power surges. As part of our new flexible framework, we propose multiple power capping policies with different degrees of awareness of fuel cell and workload behavior, and evaluate their impact on workload performance and ESD size. Using traces from Microsoft's production data center systems, we demonstrate that SizeCap significantly reduces the ESD size without violating any SLAs.

cs.DC↗

arXiv2018

A 3.8 ps RMS time synchronization implemented in a 20 nm FPGA

A 3.8ps root mean square (RMS) time synchronization implemented in a 20nm fabrication process ultrascale kintex Field Programmable Gate Array (FPGA) is presented. The multichannel high-speed serial transceivers (e.g. GTH) play a key role in a wide range of applications, such as the optical source for quantum key distribution systems. However, owing to the independent clock dividers existed in each transceiver, the random skew would appear among the multiple channels every time the system powers up or resets. A self-phase alignment method provided by Xilinx Corporation could reach a precision with 22 ps RMS and 100 ps maximum variation, which is far from meeting the demand of applications with rate up to 2.5 Gbps. To implement a high-precision intrachannel time synchronization, a protocol combined of a high-precision time-to-digital converter (TDC) and a tunable phase interpolator (PI) is presented. The TDC based on the carry8 primitive is applied to measure the intrachannel skew with 40.7ps bin size. The embedded tunable PI in each GTH channel has a theoretical step size of 3.125 ps. By tuning the PI in the minimal step size, the final intrachannel time synchronization reaches a 3.8 ps RMS precision and maximal variation 20 ps, much better than the self-phase alignment method. Besides, a desirable time offset of every channel can be implemented with a closed-loop control.

eess.SP↗

arXiv2018

Study of the rare earth effects on the magnetic fluctuations in RbLn$_2$Fe$_4$As$_4$O$_2$ (Ln = Tb, Dy and Ho) by Mössbauer spectroscopy

In the current work, we report the investigation of RbLn$_2$Fe$_4$As$_4$O$_2$ (Ln = Tb, Dy and Ho) and RbFe$_2$As$_2$ by $^{57}$Fe Mössbauer spectroscopy. Singlet pattern has been observed for all the samples indicating the absence of static magnetic ordering down to 5.9\,K on the Fe sublattice. The observed intermediate value of the isomer shift confirms the effective self charge transfer effect for the studied superconductors. Debye temperatures of these samples have been determined by the temperature dependence of the isomer shift. Most importantly, we observe different spectral line width broadening behaviors for samples with different rare earth elements, and no line width broadenings for samples Ln=Tb and Ho. These results suggest that the observed magnetic fluctuations at the Fe site might be sensitive to the magnetic behaviors of the rare earth magnetic moments.

cond-mat.str-el↗

arXiv2018

Progressive Neural Networks for Image Classification

The inference structures and computational complexity of existing deep neural networks, once trained, are fixed and remain the same for all test images. However, in practice, it is highly desirable to establish a progressive structure for deep neural networks which is able to adapt its inference process and complexity for images with different visual recognition complexity. In this work, we develop a multi-stage progressive structure with integrated confidence analysis and decision policy learning for deep neural networks. This new framework consists of a set of network units to be activated in a sequential manner with progressively increased complexity and visual recognition power. Our extensive experimental results on the CIFAR-10 and ImageNet datasets demonstrate that the proposed progressive deep neural network is able to obtain more than 10 fold complexity scalability while achieving the state-of-the-art performance using a single network model satisfying different complexity-accuracy requirements.

cs.CV↗

arXiv2018

Hadron Spectra, Decays and Scattering Properties within Basis Light Front Quantization

We survey recent progress in calculating properties of the electron and hadrons within the Basis Light Front Quantization (BLFQ) approach. We include applications to electromagnetic and strong scattering processes in relativistic heavy ion collisions. We present an initial investigation into the glueball states by applying BLFQ with multigluon sectors, introducing future research possibilities on multi-quark and multi-gluon systems.

nucl-th↗

arXiv2018

Quantum Electromagnetic Stress Tensor in an Inhomogeneous Medium

Continuing a program of examining the behavior of the vacuum expectation value of the stress tensor in a background which varies only in a single direction, we here study the electromagnetic stress tensor in a medium with permittivity depending on a single spatial coordinate, specifically, a planar dielectric half-space facing a vacuum region. There are divergences occurring that are regulated by temporal and spatial point-splitting, which have a universal character for both transverse electric and transverse magnetic modes. The nature of the divergences depends on the model of dispersion adopted. And there are singularities occurring at the edge between the dielectric and vacuum regions, which also have a universal character, depending on the structure of the discontinuities in the material properties there. Remarks are offered concerning renormalization of such models, and the significance of the stress tensor. The ambiguity in separating "bulk" and "scattering" parts of the stress tensor is discussed.

hep-th↗

arXiv2018

Coulomb Excitation of Deuteron in Peripheral Collisions with a Heavy Ion

We develop an ab initio, non-perturbative, time-dependent Basis Function (tBF) method to solve the nuclear structure and scattering problems in a unified manner. We apply this method to a test problem: the Coulomb excitation of a trapped deuteron by an impinging heavy ion. The states of the deuteron system are obtained by the ab initio nuclear structure calculation implementing a realistic inter-nucleon interaction with a weak external trap to localize the center of mass and to discretize the continuum. The evolution of the internal state of the deuteron system is directly solved using the equation of motion for the scattering. We analyze the excitation mechanism of the deuteron system by investigating its internal transition probabilities and observables as functions of the exposure time and the incident speed. In this investigation, the dynamics of the Coulomb excitation are revealed by the time evolution of the system's internal charge distribution.

nucl-th↗

arXiv2018

Frame dependence of form factors in light-front dynamics

In light-front dynamics, form factors are traditionally computed with the "good current" $J^+$ within the Drell-Yan frame $q^+=0$. Due to truncations imposed in practical calculations, the from factor may acquire frame dependence, which is often neglected. In this work, we explore the form factors in more general frames, preserving the boost covariance. We find the frame dependence of the elastic form factors for mesons is small in basis light-front holography and related models with two-body Fock space truncation. We suggest to use the difference between form factor results from Drell-Yan frame and the "longitudinal frame" as a metric for the violation of the Lorentz symmetry due to Fock space truncation.

hep-ph↗

arXiv2018

Radiative transitions between $0^{-+}$ and $1^{--}$ heavy quarkonia on the light front

We present calculations of radiative transitions between vector and pseudoscalar quarkonia in the light-front Hamiltonian approach. The valence sector light-front wavefunctions of heavy quarkonia are obtained from the Basis Light-Front Quantization (BLFQ) approach in a holographic basis. We study the transition form factor with both the traditional "good current" $J^+$ and the transverse current $\vec J_\perp$ (in particular, $J^R=J^x+i J^y$). This allows us to investigate the role of rotational symmetry by considering vector mesons with different magnetic projections ($m_j=0,\pm 1$). We use the $m_j=0$ state of the vector meson to obtain the transition form factor, since this procedure employs the dominant spin components of the light-front wavefunctions and is more robust in practical calculations. While the $m_j=\pm 1$ states are also examined, transition form factors depend on subdominant components of the light-front wavefunctions and are less robust. Transitions between states below the open-flavor thresholds are computed, including those for excited states. Comparisons are made with the experimental measurements as well as with Lattice QCD and quark model results. In addition, we apply the transverse current to calculate the decay constant of vector mesons where we obtain consistent results using either $m_j=0$ or $m_j=1$ light-front wavefunctions. This consistency provides evidence for features of rotational symmetry within the model.

hep-ph↗

arXiv2018

Tensor graph convolutional neural network

In this paper, we propose a novel tensor graph convolutional neural network (TGCNN) to conduct convolution on factorizable graphs, for which here two types of problems are focused, one is sequential dynamic graphs and the other is cross-attribute graphs. Especially, we propose a graph preserving layer to memorize salient nodes of those factorized subgraphs, i.e. cross graph convolution and graph pooling. For cross graph convolution, a parameterized Kronecker sum operation is proposed to generate a conjunctive adjacency matrix characterizing the relationship between every pair of nodes across two subgraphs. Taking this operation, then general graph convolution may be efficiently performed followed by the composition of small matrices, which thus reduces high memory and computational burden. Encapsuling sequence graphs into a recursive learning, the dynamics of graphs can be efficiently encoded as well as the spatial layout of graphs. To validate the proposed TGCNN, experiments are conducted on skeleton action datasets as well as matrix completion dataset. The experiment results demonstrate that our method can achieve more competitive performance with the state-of-the-art methods.

cs.CV↗

arXiv2018

Predicting Human Performance in Vertical Menu Selection Using Deep Learning

Predicting human performance in interaction tasks allows designers or developers to understand the expected performance of a target interface without actually testing it with real users. In this work, we present a deep neural net to model and predict human performance in performing a sequence of UI tasks. In particular, we focus on a dominant class of tasks, i.e., target selection from a vertical list or menu. We experimented with our deep neural net using a public dataset collected from a desktop laboratory environment and a dataset collected from hundreds of touchscreen smartphone users via crowdsourcing. Our model significantly outperformed previous methods on these datasets. Importantly, our method, as a deep model, can easily incorporate additional UI attributes such as visual appearance and content semantics without changing model architectures. By understanding about how a deep learning model learns from human behaviors, our approach can be seen as a vehicle to discover new patterns about human behaviors to advance analytical modeling.

cs.HC↗

arXiv2018

Mössbauer spectroscopy study of magnetic fluctuations in superconducting RbGd$_2$Fe$_4$As$_4$O$_2$

$^{57}$Fe Mössbauer spectra were measured at different temperatures between 5.9 K and 300 K on the recently discovered self-doped superconducting RbGd$_2$Fe$_4$As$_4$O$_2$ with T$_c$ as high as 35 K. Singlet pattern was observed down to the lowest temperature measured in this work, indicating the absence of static magnetic order on the Fe site. The intermediate isomer shift in comparison with that of the samples RbFe$_2$As$_2$ and GdFeAsO confirms the self doping induced local electronic structure change. Surprisingly, we observe two magnetic fluctuation induced spectral broadenings below $\sim$15 K and $\sim$100 K which are believed to be originated from the transferred magnetic fluctuations of the Gd$^{3+}$ moments and that of the magnetic fluctuations of the Fe atoms, respectively.

cond-mat.supr-con↗

arXiv2018

Nonlinear Online Learning with Adaptive Nyström Approximation

Use of nonlinear feature maps via kernel approximation has led to success in many online learning tasks. As a popular kernel approximation method, Nyström approximation, has been well investigated, and various landmark points selection methods have been proposed to improve the approximation quality. However, these improved Nyström methods cannot be directly applied to the online learning setting as they need to access the entire dataset to learn the landmark points, while we need to update model on-the-fly in the online setting. To address this challenge, we propose Adaptive Nyström approximation for solving nonlinear online learning problems. The key idea is to adaptively modify the landmark points via online kmeans and adjust the model accordingly via solving least square problem followed by a gradient descent step. We show that the resulting algorithm outperforms state-of-the-art online learning methods under the same budget.

cs.LG↗