arXiv subjects

Wei Li

Publications and source records attributed to Wei Li.

At least 1,333 records · Page 74Linked to original sources

arXiv2017

EAC-Net: A Region-based Deep Enhancing and Cropping Approach for Facial Action Unit Detection

In this paper, we propose a deep learning based approach for facial action unit detection by enhancing and cropping the regions of interest. The approach is implemented by adding two novel nets (layers): the enhancing layers and the cropping layers, to a pretrained CNN model. For the enhancing layers, we designed an attention map based on facial landmark features and applied it to a pretrained neural network to conduct enhanced learning (The E-Net). For the cropping layers, we crop facial regions around the detected landmarks and design convolutional layers to learn deeper features for each facial region (C-Net). We then fuse the E-Net and the C-Net to obtain our Enhancing and Cropping (EAC) Net, which can learn both feature enhancing and region cropping functions. Our approach shows significant improvement in performance compared to the state-of-the-art methods applied to BP4D and DISFA AU datasets.

cs.CV↗

arXiv2017

From Formalised State Machines to Implementations of Robotic Controllers

Controllers for autonomous robotic systems can be specified using state machines. However, these are typically developed in an ad hoc manner without formal semantics, which makes it difficult to analyse the controller. Simulations are often used during the development, but a rigorous connection between the designed controller and the implementation is often overlooked. This paper presents a state-machine based notation, RoboChart, together with a tool to automatically create code from the state machines, establishing a rigorous connection between specification and implementation. In RoboChart, a robot's controller is specified either graphically or using a textual description language. The controller code for simulation is automatically generated through a direct mapping from the specification. We demonstrate our approach using two case studies (self-organized aggregation and swarm taxis) in swarm robotics. The simulations are presented using two different simulators showing the general applicability of our approach.

cs.RO↗

arXiv2016

Fast color transfer from multiple images

Color transfer between images uses the statistics information of image effectively. We present a novel approach of local color transfer between images based on the simple statistics and locally linear embedding. A sketching interface is proposed for quickly and easily specifying the color correspondences between target and source image. The user can specify the correspondences of local region using scribes, which more accurately transfers the target color to the source image while smoothly preserving the boundaries, and exhibits more natural output results. Our algorithm is not restricted to one-to-one image color transfer and can make use of more than one target images to transfer the color in different regions in the source image. Moreover, our algorithm does not require to choose the same color style and image size between source and target images. We propose the sub-sampling to reduce the computational load. Comparing with other approaches, our algorithm is much better in color blending in the input data. Our approach preserves the other color details in the source image. Various experimental results show that our approach specifies the correspondences of local color region in source and target images. And it expresses the intention of users and generates more actual and natural results of visual effect.

cs.CV↗

arXiv2016

Multi-resolution community detection in massive networks

Aiming at improving the efficiency and accuracy of community detection in complex networks, we proposed a new algorithm, which is based on the idea that communities could be detected from subnetworks by comparing the internal and external cohesion of each subnetwork. In our method, similar nodes are firstly gathered into meta-communities, which are then decided to be retained or merged through a multilevel label propagation process, until all of them meet our community criterion. Our algorithm requires neither any priori information of communities nor optimization of any objective function. Experimental results on both synthetic and real-world networks show that, our algorithm performs quite well and runs extremely fast, compared with several other popular algorithms. By tuning a resolution parameter, we can also observe communities at different scales, so this could reveal the hierarchical structure of the network. To further explore the effectiveness of our method, we applied it to the E-Coli transcriptional regulatory network, and found that all the identified modules have strong structural and functional coherence.

physics.soc-ph↗

arXiv2016

Fault Diagnosis of Rolling Element Bearings with a Spectrum Searching Method

Rolling element bearing faults in rotating systems are observed as impulses in the vibration signals, which are usually buried in noises. In order to effectively detect the fault of bearings, a novel spectrum searching method is proposed. The structural information of spectrum (SIOS) on a predefined basis is constructed through a searching algorithm, such that the harmonics of impulses generated by faults can be clearly identified and analyzed. Local peaks of the spectrum are located on a certain bin of the basis, and then the SIOS can interpret the spectrum via the number and energy of harmonics related to frequency bins of the basis. Finally bearings can be diagnosed based on the SIOS by identifying its dominant components. Mathematical formulation is developed to guarantee the correct construction of the SISO through searching. The effectiveness of the proposed method is verified with a simulation signal and a benchmark study of bearings.

cs.SD↗

arXiv2016

Chinese Poetry Generation with Planning based Neural Network

Chinese poetry generation is a very challenging task in natural language processing. In this paper, we propose a novel two-stage poetry generating method which first plans the sub-topics of the poem according to the user's writing intent, and then generates each line of the poem sequentially, using a modified recurrent neural network encoder-decoder framework. The proposed planning-based method can ensure that the generated poem is coherent and semantically consistent with the user's intent. A comprehensive evaluation with human judgments demonstrates that our proposed approach outperforms the state-of-the-art poetry generating methods and the poem quality is somehow comparable to human poets.

cs.CL↗

arXiv2016

Bilayer Linearized Tensor Renormalization Group Approach for Thermal Tensor Networks

In this paper, we perform a comprehensive study of the renormalization group (RG) method on thermal tensor networks (TTN). By Trotter-Suzuki decomposition, one obtains the 1+1D TTN representing the partition function of 1D quantum lattice models, and then employs efficient RG contractions to obtain the thermodynamic properties with high precision. The linearized tensor renormalization group (LTRG) method, which can be used to contract TTN in an efficient and accurate way, is briefly reviewed. In addition, the single-layer LTRG can be generalized to a bilayer form, dubbed as LTRG++, in both finite- and infinite-size systems, with accuracies significantly improved. We provide the details of LTRG++ in finite-size system, comparing its accuracy with single-layer algorithm, and elaborate the infinite-size LTRG++ in the context of fermion chain model. We show that the LTRG++ algorithm in infinite-size system is in essence equivalent to transfer-matrix renormalization group method, while expressed in a tensor network language. LTRG++ is then applied to study an extended Hubbard model, where the phase separation phenomenon, groundstate phase diagram, as well as quantum criticality-enhanced magnetocaloric effects under external fields, are investigated.

cond-mat.str-el↗

arXiv2016

Experimental ten-photon entanglement

Quantum entanglement among multiple spatially separated particles is of fundamental interest, and can serve as central resources for studies in quantum nonlocality, quantum-to-classical transition, quantum error correction, and quantum simulation. The ability of generating an increasing number of entangled particles is an important benchmark for quantum information processing. The largest entangled states were previously created with fourteen trapped ions, eight photons, and five superconducting qubits. Here, based on spontaneous parametric down-converted two-photon entanglement source with simultaneously a high brightness of ~12 MHz/W, a collection efficiency of ~70% and an indistinguishability of ~91% between independent photons, we demonstrate, for the first time, genuine and distillable entanglement of ten single photons under different pump power. Our work creates a state-of-the-art platform for multi-photon experiments, and provide enabling technologies for challenging optical quantum information tasks such as high-efficiency scattershot boson sampling with many photons.

quant-ph↗

arXiv2016

Second-Harmonic Imaging in Random Media

We consider the problem of optical imaging of small nonlinear scatterers in random media. We propose an extension of coherent interferometric imaging (CINT) that applies to scatterers that emit second-harmonic light. We compare this method to a nonlinear version of migration imaging and find that the images obtained by CINT are more robust to statistical fluctuations. This finding is supported by a resolution analysis that is carried out in the setting of geometrical optics in random media. It is also consistent with numerical simulations for which the assumptions of the geometrical optics model do not hold.

math.NA↗

arXiv2016

Community detection by label propagation with compression of flow

The label propagation algorithm (LPA) has been proved to be a fast and effective method for detecting communities in large complex networks. However, its performance is subject to the non-stable and trivial solutions of the problem. In this paper, we propose a modified label propagation algorithm LPAf to efficiently detect community structures in networks. Instead of the majority voting rule of the basic LPA, LPAf updates the label of a node by considering the compression of a description of random walks on a network. A multi-step greedy agglomerative strategy is employed to enable LPAf to escape the local optimum. Furthermore, an incomplete update condition is also adopted to speed up the convergence. Experimental results on both synthetic and real-world networks confirm the effectiveness of our algorithm.

physics.soc-ph↗

arXiv2016

Recurrent Neural Network Language Model Adaptation Derived Document Vector

In many natural language processing (NLP) tasks, a document is commonly modeled as a bag of words using the term frequency-inverse document frequency (TF-IDF) vector. One major shortcoming of the frequency-based TF-IDF feature vector is that it ignores word orders that carry syntactic and semantic relationships among the words in a document, and they can be important in some NLP tasks such as genre classification. This paper proposes a novel distributed vector representation of a document: a simple recurrent-neural-network language model (RNN-LM) or a long short-term memory RNN language model (LSTM-LM) is first created from all documents in a task; some of the LM parameters are then adapted by each document, and the adapted parameters are vectorized to represent the document. The new document vectors are labeled as DV-RNN and DV-LSTM respectively. We believe that our new document vectors can capture some high-level sequential information in the documents, which other current document representations fail to capture. The new document vectors were evaluated in the genre classification of documents in three corpora: the Brown Corpus, the BNC Baby Corpus and an artificially created Penn Treebank dataset. Their classification performances are compared with the performance of TF-IDF vector and the state-of-the-art distributed memory model of paragraph vector (PV-DM). The results show that DV-LSTM significantly outperforms TF-IDF and PV-DM in most cases, and combinations of the proposed document vectors with TF-IDF or PV-DM may further improve performance.

cs.CL↗

arXiv2016

Learning and Fusing Multimodal Features from and for Multi-task Facial Computing

We propose a deep learning-based feature fusion approach for facial computing including face recognition as well as gender, race and age detection. Instead of training a single classifier on face images to classify them based on the features of the person whose face appears in the image, we first train four different classifiers for classifying face images based on race, age, gender and identification (ID). Multi-task features are then extracted from the trained models and cross-task-feature training is conducted which shows the value of fusing multimodal features extracted from multi-tasks. We have found that features trained for one task can be used for other related tasks. More interestingly, the features trained for a task with more classes (e.g. ID) and then used in another task with fewer classes (e.g. race) outperforms the features trained for the other task itself. The final feature fusion is performed by combining the four types of features extracted from the images by the four classifiers. The feature fusion approach improves the classifications accuracy by a 7.2%, 20.1%, 22.2%, 21.8% margin, respectively, for ID, age, race and gender recognition, over the results of single classifiers trained only on their individual features. The proposed method can be applied to applications in which different types of data or features can be extracted.

cs.CV↗

arXiv2016

Optimized motor imagery paradigm based on imagining Chinese characters writing movement

Motor imagery (MI) is a mental representation of motor behavior that has been widely used as a control method for a brain-computer interface (BCI), allowing communication for the physically impaired. The performance of MI based BCI mainly depends on the subject's ability to self-modulate EEG signals. Proper training can help naive subjects learn to modulate brain activity proficiently. However, training subjects typically involves abstract motor tasks and is time-consuming. To improve the performance of naive subjects during motor imagery, a novel paradigm was presented that would guide naive subjects to modulate brain activity effectively. In this new paradigm, pictures of the left or right hand were used as cues for subjects to finish the motor imagery task. Fourteen healthy subjects (11 male, aged 22-25 years, mean 23.6+/-1.16) participated in this study. The task was to imagine writing a Chinese character. Specifically, subjects could imagine hand movements following the sequence of writing strokes in the Chinese character. This paradigm was meant to find an effective and familiar action for most Chinese people, to provide them with a specific, extensively practiced task and help them modulate brain activity. Results showed that the writing task paradigm yielded significantly better performance than the traditional arrow paradigm (p<0.001). Questionnaire replies indicated that most subjects thought the new paradigm was easier and more comfortable. The proposed new motor imagery paradigm could guide subjects to help them modulate brain activity effectively. Results showed that there were significant improvements using new paradigm, both in classification accuracy and usability.

cs.HC↗

arXiv2016

Turing learning: a metric-free approach to inferring behavior and its application to swarms

We propose Turing Learning, a novel system identification method for inferring the behavior of natural or artificial systems. Turing Learning simultaneously optimizes two populations of computer programs, one representing models of the behavior of the system under investigation, and the other representing classifiers. By observing the behavior of the system as well as the behaviors produced by the models, two sets of data samples are obtained. The classifiers are rewarded for discriminating between these two sets, that is, for correctly categorizing data samples as either genuine or counterfeit. Conversely, the models are rewarded for 'tricking' the classifiers into categorizing their data samples as genuine. Unlike other methods for system identification, Turing Learning does not require predefined metrics to quantify the difference between the system and its models. We present two case studies with swarms of simulated robots and prove that the underlying behaviors cannot be inferred by a metric-based system identification method. By contrast, Turing Learning infers the behaviors with high accuracy. It also produces a useful by-product - the classifiers - that can be used to detect abnormal behavior in the swarm. Moreover, we show that Turing Learning also successfully infers the behavior of physical robot swarms. The results show that collective behaviors can be directly inferred from motion trajectories of individuals in the swarm, which may have significant implications for the study of animal collectives. Furthermore, Turing Learning could prove useful whenever a behavior is not easily characterizable using metrics, making it suitable for a wide range of applications.

stat.ML↗

arXiv2016

Criticality in Two-Dimensional Quantum Systems: Tensor Network Approach

Determination and characterization of criticality in two-dimensional (2D) quantum many-body systems belong to the most important challenges and problems of quantum physics. In this paper we propose an efficient scheme to solve this problem by utilizing the infinite projected entangled pair state (iPEPS), and tensor network (TN) representations. We show that the criticality of a 2D state is faithfully reproduced by the ground state (dubbed as boundary state) of a one-dimensional effective Hamiltonian constructed from its iPEPS representation. We demonstrate that for a critical state the correlation length and the entanglement spectrum of the boundary state are essentially different from those of a gapped iPEPS. This provides a solid indicator that allows to identify the criticality of the 2D state. Our scheme is verified on the resonating valence bond (RVB) states on kagomé and square lattices, where the boundary state of the honeycomb RVB is found to be described by a $c=1$ conformal field theory. We apply our scheme also to the ground state of the spin-1/2 XXZ model on honeycomb lattice, illustrating the difficulties of standard variational TN approaches to study the critical ground states. Our scheme is of high versatility and flexibility, and can be further applied to investigate the quantum criticality in many other phenomena, such as finite-temperature and topological phase transitions.

cond-mat.str-el↗

arXiv2016

Spin ordered ground state and thermodynamic behaviors of the spin-3/2 kagome Heisenberg antiferromagnet

Three different tensor network optimization algorithms are employed to accurately determine the ground state and thermodynamic properties of the spin-3/2 kagome Heisenberg antiferromagnet. We found that the $\sqrt{3} \times \sqrt{3}$ state, rather than the $q = 0$ state, is the ground state of this system, and such an ordered state is melted at any finite temperature, thereby clarifying the existing experimental controversies. A 1/3-magnetization plateau in the magnetic curve is observed, being consistent with the experimental observation. The absence of a zero-magnetization plateau indicates a gapless spin excitation that is further supported by the thermodynamic asymptotic behaviors of the susceptibility and specific heat. At low temperatures, the specific heat is shown to exhibit a $\sqrt{T}$ behavior, and the susceptibility approaches a finite constant as $T\rightarrow 0$. Our TN results of thermodynamic properties are compared with those from high temperature series expansion. In addition, we observe a quantum phase transition between $q = 0$ and $\sqrt{3}\times\sqrt{3}$ states in a spin-3/2 kagome XXZ model at the critical point $Δ_c = 0.408$. This study provides reliable and useful information for further explorations on high spin kagome physics.

cond-mat.str-el↗

arXiv2016

The effects of overtaking strategy in the Nagel-Schreckenberg model

Based on the Nagel-Schreckenberg (NS) model with periodic boundary conditions, we proposed the NSOS model by adding the overtaking strategy (OS). In our model, overtaking vehicles are randomly selected with probability $q$ at each time step, and the successful overtaking is determined by their velocities. We observed that (i) traffic jams still occur in the NSOS model; (ii) OS increases the traffic flow in the regime where the densities exceed the maximum flow density. We also studied the phase transition (from free flow phase to jammed phase) of the NSOS model by analyzing the overtaking success rate, order parameter, relaxation time and correlation function, respectively. It was shown that the NSOS model differs from the NS model mainly in the jammed regime, and the influence of OS on the transition density is dominated by the braking probability $p$

nlin.CG↗

arXiv2016

Differential Chow varieties exist

Chow varieties are a parameter space for cycles of a given variety of a given codimension and degree. We construct their analog for differential algebraic varieties with differential algebraic subvarieties, answering a question of Gao, Li and Wei. The proof uses the construction of classical algebro-geometric Chow varieties, the theory of characteristic sets of differential varieties, the theory of prolongation spaces, and the theory of differential Chow forms. In the course of the proof several definability results from the theory of algebraically closed fields are required. Elementary proofs of these results are given in an appendix by William Johnson.

math.AG↗