arXiv subjects

Yi Chen

Publications and source records attributed to Yi Chen.

302 records · Page 17Linked to original sources

arXiv2014

Linearly-Coupled Fountain Codes

Network-coded multiple access (NCMA) is a communication scheme for wireless multiple-access networks where physical-layer network coding (PNC) is employed. In NCMA, a user encodes and spreads its message into multiple packets. Time is slotted and multiple users transmit packets (one packet each) simultaneously in each timeslot. A sink node aims to decode the messages of all the users from the sequence of receptions over successive timeslots. For each timeslot, the NCMA receiver recovers multiple linear combinations of the packets transmitted in that timeslot, forming a system of linear equations. Different systems of linear equations are recovered in different timeslots. A message decoder then recovers the original messages of all the users by jointly solving multiple systems of linear equations obtained over different timeslots. We propose a low-complexity digital fountain approach for this coding problem, where each source node encodes its message into a sequence of packets using a fountain code. The aforementioned systems of linear equations recovered by the NCMA receiver effectively couple these fountain codes together. We refer to the coupling of the fountain codes as a linearly-coupled (LC) fountain code. The ordinary belief propagation (BP) decoding algorithm for conventional fountain codes is not optimal for LC fountain codes. We propose a batched BP decoding algorithm and analyze the convergence of the algorithm for general LC fountain codes. We demonstrate how to optimize the degree distributions and show by numerical results that the achievable rate region is nearly optimal. Our approach significantly reduces the decoding complexity compared with the previous NCMA schemes based on Reed-Solomon codes and random linear codes, and hence has the potential to increase throughput and decrease delay in computation-limited NCMA systems.

cs.IT↗

arXiv2014

Observation of vacancy-induced suppression of electronic cooling in defected graphene

Previous studies of electron-phonon interaction in impure graphene have found that static disorder can give rise to an enhancement of electronic cooling. We investigate the effect of dynamic disorder and observe over an order of magnitude suppression of electronic cooling compared with clean graphene. The effect is stronger in graphene with more vacancies, confirming its vacancy-induced nature. The dependence of the coupling constant on the phonon temperature implies its link to the dynamics of disorder. Our study highlights the effect of disorder on electron-phonon interaction in graphene. In addition, the suppression of electronic cooling holds great promise for improving the performance of graphene-based bolometer and photo-detector devices.

cond-mat.mes-hall↗

arXiv2014

New Observables for CP Violation in Higgs Decays

Current experimental data on the 125 GeV Higgs boson still allow room for large CP violation. The observables usually considered in this context are triple product asymmetries, which require an input of four visible particles after imposing momentum conservation. We point out a new class of CP violating observables in Higgs physics which require only three reconstructed momenta. They may arise if the process involves an interference of amplitudes with different intermediate particles, which provide distinct strong phases in the form of the Breit-Wigner widths, in addition to possible weak phases that arise from CP violating couplings of the Higgs in the Lagrangian. As an example, we propose a forward-backward asymmetry of the charged lepton in the three-body Higgs decay, $h\to \ell^-\ell^+ γ$, as a probe for CP-violating Higgs couplings to $Zγ$ and $γγ$ pairs. Other processes exhibiting this type of CP-violation are also discussed.

hep-ph↗

arXiv2014

Overlapping community detection in signed networks

Complex networks considering both positive and negative links have gained considerable attention during the past several years. Community detection is one of the main challenges for complex network analysis. Most of the existing algorithms for community detection in a signed network aim at providing a hard-partition of the network where any node should belong to a community or not. However, they cannot detect overlapping communities where a node is allowed to belong to multiple communities. The overlapping communities widely exist in many real world networks. In this paper, we propose a signed probabilistic mixture (SPM) model for overlapping community detection in signed networks. Compared with the existing models, the advantages of our methodology are (i) providing soft-partition solutions for signed networks; (ii) providing soft-memberships of nodes. Experiments on a number of signed networks show that our SPM model: (i) can identify assortative structures or disassortative structures as the same as other state-of-the-art models; (ii) can detect overlapping communities; (iii) outperform other state-of-the-art models at shedding light on the community detection in synthetic signed networks.

cs.SI↗

arXiv2014

Extracting Effective Higgs Couplings in the Golden Channel

Kinematic distributions in Higgs decays to four charged leptons, the so called `golden channel, are a powerful probe of the tensor structure of its couplings to neutral electroweak gauge bosons. In this study we construct the first part of a comprehensive analysis framework designed to maximize the information contained in this channel in order to perform direct extraction of the various possible Higgs couplings. To that end we first complete an earlier analytic calculation of the leading order fully differential cross sections for the golden channel signal and background to include the 4e and $4μ$ final states with interference between identical final states. We also examine the relative fractions of the different possible combinations of scalar-tensor couplings by integrating the fully differential cross section over all kinematic variables as well as show various doubly differential spectra for both the signal and background. From these analytic expressions we then construct a `generator level' analysis framework based on the maximum likelihood method. We demonstrate the ability of our framework to perform multi-parameter extractions of all the possible effective couplings of a spin-0 scalar to pairs of neutral electroweak gauge bosons including any correlations. This framework provides a powerful method for study of these couplings and can be readily adapted to include the relevant detector and systematic effects which we demonstrate in an accompanying study to follow.

hep-ph↗

arXiv2014

Global Solutions to Nonconvex Optimization of 4th-Order Polynomial and Log-Sum-Exp Functions

This paper presents a canonical dual approach for solving a nonconvex global optimization problem governed by a sum of fourth-order polynomial and a log-sum-exp function. Such a problem arises extensively in engineering and sciences. Based on the canonical duality-triality theory, this nonconvex problem is transformed to an equivalent dual problem, which can be solved easily under certain conditions. We proved that both global minimizer and the biggest local extrema of the primal problem can be obtained analytically from the canonical dual solutions. As two special cases, a quartic polynomial minimization and a minimax problem are discussed. Existence conditions are derived, which can be used to classify easy and relative hard instances. Applications are illustrated by several nonconvex and nonsmooth examples.

math.OC↗

arXiv2013

Scrutinizing the Higgs Signal and Background in the $2e2μ$ Golden Channel

Kinematic distributions in the decays of the newly discovered resonance to four leptons are a powerful test of the tensor structure of its couplings to electroweak gauge bosons. We present an analytic calculation for both signal and background of the fully differential cross section for the `Golden Channel' $e^+e^-μ^+μ^-$ final state. We include all interference effects between intermediate gauge bosons and allow them to be on- or off-shell. For the signal we compute the fully differential cross section for general scalar couplings to $ZZ$, $γγ$, and $Zγ$. For the background we compute the leading order fully differential cross section for the dominant contribution coming from $q\bar{q}$ annihilation into $Z$ and $γ$ gauge bosons, including the contribution from the resonant $Z\rightarrow 2e2μ$ process. We also present singly and doubly differential projections and study the interference effects on the differential spectra. These expressions can be used in a variety of ways to uncover the nature of the newly discovered resonance or any new scalars decaying to neutral gauge bosons which might be discovered in the future.

hep-ph↗

arXiv2013

Global Solutions to Large-Scale Spherical Constrained Quadratic Minimization via Canonical Dual Approach

This paper presents global optimal solutions to a nonconvex quadratic minimization problem over a sphere constraint. The problem is well-known as a trust region subproblem and has been studied extensively for decades. The main challenge is the so called 'hard case', i.e., the problem has multiple solutions on the boundary of the sphere. By canonical duality theory, this challenging problem is able to reformed as an one-dimensional canonical dual problem without duality gap. Sufficient and necessary conditions are obtained by the triality theory, which can be used to identify whether the problem is hard case or not. A perturbation method and the associated algorithms are proposed to solve this hard case problem. Theoretical results and methods are verified by large-size examples.

math.OC↗

arXiv2013

Bio-inspired data mining: Treating malware signatures as biosequences

The application of machine learning to bioinformatics problems is well established. Less well understood is the application of bioinformatics techniques to machine learning and, in particular, the representation of non-biological data as biosequences. The aim of this paper is to explore the effects of giving amino acid representation to problematic machine learning data and to evaluate the benefits of supplementing traditional machine learning with bioinformatics tools and techniques. The signatures of 60 computer viruses and 60 computer worms were converted into amino acid representations and first multiply aligned separately to identify conserved regions across different families within each class (virus and worm). This was followed by a second alignment of all 120 aligned signatures together so that non-conserved regions were identified prior to input to a number of machine learning techniques. Differences in length between virus and worm signatures after the first alignment were resolved by the second alignment. Our first set of experiments indicates that representing computer malware signatures as amino acid sequences followed by alignment leads to greater classification and prediction accuracy. Our second set of experiments indicates that checking the results of data mining from artificial virus and worm data against known proteins can lead to generalizations being made from the domain of naturally occurring proteins to malware signatures. However, further work is needed to determine the advantages and disadvantages of different representations and sequence alignment methods for handling problematic machine learning data.

cs.LG↗

arXiv2012

Bayesian Data Cleaning for Web Data

Data Cleaning is a long standing problem, which is growing in importance with the mass of uncurated web data. State of the art approaches for handling inconsistent data are systems that learn and use conditional functional dependencies (CFDs) to rectify data. These methods learn data patterns--CFDs--from a clean sample of the data and use them to rectify the dirty/inconsistent data. While getting a clean training sample is feasible in enterprise data scenarios, it is infeasible in web databases where there is no separate curated data. CFD based methods are unfortunately particularly sensitive to noise; we will empirically demonstrate that the number of CFDs learned falls quite drastically with even a small amount of noise. In order to overcome this limitation, we propose a fully probabilistic framework for cleaning data. Our approach involves learning both the generative and error (corruption) models of the data and using them to clean the data. For generative models, we learn Bayes networks from the data. For error models, we consider a maximum entropy framework for combing multiple error processes. The generative and error models are learned directly from the noisy data. We present the details of the framework and demonstrate its effectiveness in rectifying web data.

cs.DB↗

arXiv2011

Discriminative Local Sparse Representations for Robust Face Recognition

A key recent advance in face recognition models a test face image as a sparse linear combination of a set of training face images. The resulting sparse representations have been shown to possess robustness against a variety of distortions like random pixel corruption, occlusion and disguise. This approach however makes the restrictive (in many scenarios) assumption that test faces must be perfectly aligned (or registered) to the training data prior to classification. In this paper, we propose a simple yet robust local block-based sparsity model, using adaptively-constructed dictionaries from local features in the training data, to overcome this misalignment problem. Our approach is inspired by human perception: we analyze a series of local discriminative features and combine them to arrive at the final classification decision. We propose a probabilistic graphical model framework to explicitly mine the conditional dependencies between these distinct sparse local features. In particular, we learn discriminative graphs on sparse representations obtained from distinct local slices of a face. Conditional correlations between these sparse features are first discovered (in the training phase), and subsequently exploited to bring about significant improvements in recognition rates. Experimental results obtained on benchmark face databases demonstrate the effectiveness of the proposed algorithms in the presence of multiple registration errors (such as translation, rotation, and scaling) as well as under variations of pose and illumination.

cs.CV↗

arXiv2011

Query Expansion Based on Clustered Results

Query expansion is a functionality of search engines that suggests a set of related queries for a user-issued keyword query. Typical corpus-driven keyword query expansion approaches return popular words in the results as expanded queries. Using these approaches, the expanded queries may correspond to a subset of possible query semantics, and thus miss relevant results. To handle ambiguous queries and exploratory queries, whose result relevance is difficult to judge, we propose a new framework for keyword query expansion: we start with clustering the results according to user specified granularity, and then generate expanded queries, such that one expanded query is generated for each cluster whose result set should ideally be the corresponding cluster. We formalize this problem and show its APX-hardness. Then we propose two efficient algorithms named iterative single-keyword refinement and partial elimination based convergence, respectively, which effectively generate a set of expanded queries from clustered results that provide a classification of the original query results. We believe our study of generating an optimal query based on the ground truth of the query results not only has applications in query expansion, but has significance for studying keyword search quality in general.

cs.IR↗

arXiv2000

Charge-Reversal Instability in Mixed Bilayer Vesicles

Bilayer vesicles form readily from mixtures of charged and neutral surfactants. When such a mixed vesicle binds an oppositely-charged object, its membrane partially demixes: the adhesion zone recruits more charged surfactants from the rest of the membrane. Given an unlimited supply of adhering objects one might expect the vesicle to remain attractive until it was completely covered. Contrary to this expectation, we show that a vesicle can instead exhibit {\it adhesion saturation,} partitioning spontaneously into an attractive zone with definite area fraction, and a repulsive zone. The latter zone rejects additional incoming objects because counterions on the interior of the vesicle migrate there, effectively reversing the membrane's charge. The effect is strongest at high surface charge densities, low ionic strength, and with thin, impermeable membranes. Adhesion saturation in such a situation has recently been observed experimentally [H. Aranda-Espinoza {\it et al.}, {\sl Science} {\bf285} 394--397 (1999)].

cond-mat.soft↗

arXiv1999

Electrostatic Repulsion of Positively Charged Vesicles and Negatively Charged Objects

A positively charged, mixed bilayer vesicle in the presence of negatively charged surfaces (for example, colloidal particles) can spontaneously partition into an adhesion zone of definite area, and another zone that repels additional negative objects. Although the membrane itself has nonnegative charge in the repulsive zone, negative counterions on the interior of the vesicle spontaneously aggregate there, and present a net negative charge to the exterior. Beyond the fundamental result that oppositely charged objects can repel, our mechanism helps explain recent experiments on surfactant vesicles.

cond-mat.soft↗