arXiv subjects

Hao Tang

Publications and source records attributed to Hao Tang.

At least 451 records · Page 25Linked to original sources

arXiv2020

Quantum Go Machine

Go has long been considered as a testbed for artificial intelligence. By introducing certain quantum features, such as superposition and collapse of wavefunction, we experimentally demonstrate a quantum version of Go by using correlated photon pairs entangled in polarization degree of freedom. The total dimension of Hilbert space of the generated states grows exponentially as two players take turns to place the stones in time series. As nondeterministic and imperfect information games are more difficult to solve using nowadays technology, we excitedly find that the inherent randomness in quantum physics can bring the game nondeterministic trait, which does not exist in the classical counterpart. Some quantum resources, like coherence or entanglement, can also be encoded to represent the state of quantum stones. Adjusting the quantum resource may vary the average imperfect information (as comparison classical Go is a perfect information game) of a single game. We further verify its non-deterministic feature by showing the unpredictability of the time series data obtained from different classes of quantum state. Finally, by comparing quantum Go with a few typical games that are widely studied in artificial intelligence, we find that quantum Go can cover a wide range of game difficulties rather than a single point. Our results establish a paradigm of inventing new games with quantum-enabled difficulties by harnessing inherent quantum features and resources, and provide a versatile platform for the test of new algorithms to both classical and quantum machine learning.

quant-ph↗

arXiv2020

Cross-View Image Synthesis with Deformable Convolution and Attention Mechanism

Learning to generate natural scenes has always been a daunting task in computer vision. This is even more laborious when generating images with very different views. When the views are very different, the view fields have little overlap or objects are occluded, leading the task very challenging. In this paper, we propose to use Generative Adversarial Networks(GANs) based on a deformable convolution and attention mechanism to solve the problem of cross-view image synthesis (see Fig.1). It is difficult to understand and transform scenes appearance and semantic information from another view, thus we use deformed convolution in the U-net network to improve the network's ability to extract features of objects at different scales. Moreover, to better learn the correspondence between images from different views, we apply an attention mechanism to refine the intermediate feature map thus generating more realistic images. A large number of experiments on different size images on the Dayton dataset[1] show that our model can produce better results than state-of-the-art methods.

cs.CV↗

arXiv2020

XingGAN for Person Image Generation

We propose a novel Generative Adversarial Network (XingGAN or CrossingGAN) for person image generation tasks, i.e., translating the pose of a given person to a desired one. The proposed Xing generator consists of two generation branches that model the person's appearance and shape information, respectively. Moreover, we propose two novel blocks to effectively transfer and update the person's shape and appearance embeddings in a crossing way to mutually improve each other, which has not been considered by any other existing GAN-based image generation work. Extensive experiments on two challenging datasets, i.e., Market-1501 and DeepFashion, demonstrate that the proposed XingGAN advances the state-of-the-art performance both in terms of objective quantitative scores and subjective visual realness. The source code and trained models are available at https://github.com/Ha0Tang/XingGAN.

cs.CV↗

arXiv2020

Belief Propagation Neural Networks

Learned neural solvers have successfully been used to solve combinatorial optimization and decision problems. More general counting variants of these problems, however, are still largely solved with hand-crafted solvers. To bridge this gap, we introduce belief propagation neural networks (BPNNs), a class of parameterized operators that operate on factor graphs and generalize Belief Propagation (BP). In its strictest form, a BPNN layer (BPNN-D) is a learned iterative operator that provably maintains many of the desirable properties of BP for any choice of the parameters. Empirically, we show that by training BPNN-D learns to perform the task better than the original BP: it converges 1.7x faster on Ising models while providing tighter bounds. On challenging model counting problems, BPNNs compute estimates 100's of times faster than state-of-the-art handcrafted methods, while returning an estimate of comparable quality.

cs.LG↗

arXiv2020

When Dictionary Learning Meets Deep Learning: Deep Dictionary Learning and Coding Network for Image Recognition with Limited Data

We present a new Deep Dictionary Learning and Coding Network (DDLCN) for image recognition tasks with limited data. The proposed DDLCN has most of the standard deep learning layers (e.g., input/output, pooling, fully connected, etc.), but the fundamental convolutional layers are replaced by our proposed compound dictionary learning and coding layers. The dictionary learning learns an over-complete dictionary for input training data. At the deep coding layer, a locality constraint is added to guarantee that the activated dictionary bases are close to each other. Then the activated dictionary atoms are assembled and passed to the compound dictionary learning and coding layers. In this way, the activated atoms in the first layer can be represented by the deeper atoms in the second dictionary. Intuitively, the second dictionary is designed to learn the fine-grained components shared among the input dictionary atoms, thus a more informative and discriminative low-level representation of the dictionary atoms can be obtained. We empirically compare DDLCN with several leading dictionary learning methods and deep learning models. Experimental results on five popular datasets show that DDLCN achieves competitive results compared with state-of-the-art methods when the training data is limited. Code is available at https://github.com/Ha0Tang/DDLCN.

cs.CV↗

arXiv2020

Relevant Region Prediction for Crowd Counting

Crowd counting is a concerned and challenging task in computer vision. Existing density map based methods excessively focus on the individuals' localization which harms the crowd counting performance in highly congested scenes. In addition, the dependency between the regions of different density is also ignored. In this paper, we propose Relevant Region Prediction (RRP) for crowd counting, which consists of the Count Map and the Region Relation-Aware Module (RRAM). Each pixel in the count map represents the number of heads falling into the corresponding local area in the input image, which discards the detailed spatial information and forces the network pay more attention to counting rather than localizing individuals. Based on the Graph Convolutional Network (GCN), Region Relation-Aware Module is proposed to capture and exploit the important region dependency. The module builds a fully connected directed graph between the regions of different density where each node (region) is represented by weighted global pooled feature, and GCN is learned to map this region graph to a set of relation-aware regions representations. Experimental results on three datasets show that our method obviously outperforms other existing state-of-the-art methods.

cs.CV↗

arXiv2020

Vector-Quantized Autoregressive Predictive Coding

Autoregressive Predictive Coding (APC), as a self-supervised objective, has enjoyed success in learning representations from large amounts of unlabeled data, and the learned representations are rich for many downstream tasks. However, the connection between low self-supervised loss and strong performance in downstream tasks remains unclear. In this work, we propose Vector-Quantized Autoregressive Predictive Coding (VQ-APC), a novel model that produces quantized representations, allowing us to explicitly control the amount of information encoded in the representations. By studying a sequence of increasingly limited models, we reveal the constituents of the learned representations. In particular, we confirm the presence of information with probing tasks, while showing the absence of information with mutual information, uncovering the model's preference in preserving speech information as its capacity becomes constrained. We find that there exists a point where phonetic and speaker information are amplified to maximize a self-supervised objective. As a byproduct, the learned codes for a particular model capacity correspond well to English phones.

eess.AS↗

arXiv2020

Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation

In this work, we propose a novel Cycle In Cycle Generative Adversarial Network (C$^2$GAN) for the task of keypoint-guided image generation. The proposed C$^2$GAN is a cross-modal framework exploring a joint exploitation of the keypoint and the image data in an interactive manner. C$^2$GAN contains two different types of generators, i.e., keypoint-oriented generator and image-oriented generator. Both of them are mutually connected in an end-to-end learnable fashion and explicitly form three cycled sub-networks, i.e., one image generation cycle and two keypoint generation cycles. Each cycle not only aims at reconstructing the input domain, and also produces useful output involving in the generation of another cycle. By so doing, the cycles constrain each other implicitly, which provides complementary information from the two different modalities and brings extra supervision across cycles, thus facilitating more robust optimization of the whole network. Extensive experimental results on two publicly available datasets, i.e., Radboud Faces and Market-1501, demonstrate that our approach is effective to generate more photo-realistic images compared with state-of-the-art models.

cs.CV↗

arXiv2020

Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation

In this paper, we address the task of semantic-guided scene generation. One open challenge in scene generation is the difficulty of the generation of small objects and detailed local texture, which has been widely observed in global image-level generation methods. To tackle this issue, in this work we consider learning the scene generation in a local context, and correspondingly design a local class-specific generative network with semantic maps as a guidance, which separately constructs and learns sub-generators concentrating on the generation of different classes, and is able to provide more scene details. To learn more discriminative class-specific feature representations for the local generation, a novel classification module is also proposed. To combine the advantage of both the global image-level and the local class-specific generation, a joint generation network is designed with an attention fusion module and a dual-discriminator structure embedded. Extensive experiments on two scene image generation tasks show superior generation performance of the proposed model. The state-of-the-art results are established by large margins on both tasks and on challenging public benchmarks. The source code and trained models are available at https://github.com/Ha0Tang/LGGAN.

cs.CV↗

arXiv2020

TensorFlow Solver for Quantum PageRank in Large-Scale Networks

Google PageRank is a prevalent and useful algorithm for ranking the significance of nodes or websites in a network, and a recent quantum counterpart for PageRank algorithm has been raised to suggest a higher accuracy of ranking comparing to Google PageRank. The quantum PageRank algorithm is essentially based on quantum stochastic walks and can be expressed using Lindblad master equation, which, however, needs to solve the Kronecker products of an O(N^4) dimension and requires severely large memory and time when the number of nodes N in a network increases above 150. Here, we present an efficient solver for quantum PageRank by using the Runge-Kutta method to reduce the matrix dimension to O(N^2) and employing TensorFlow to conduct GPU parallel computing. We demonstrate its performance in solving quantum PageRank for the USA major airline network with up to 922 nodes. Compared with the previous quantum PageRank solver, our solver dramatically reduces the required memory and time to only 1% and 0.2%, respectively, making it practical to work in a normal computer with a memory of 4-8 GB in no more than 100 seconds. This efficient solver for large-scale quantum PageRank and quantum stochastic walks would greatly facilitate studies of quantum information in real-life applications.

quant-ph↗

arXiv2020

A Hybrid Quantum Memory Enabled Network at Room Temperature

Quantum memory capable of storage and retrieval of flying photons on demand is crucial for developing quantum information technologies. However, the devices needed for long-distance links are quite different from those envisioned for local processing. Here, we present the first hybrid quantum memory enabled network by demonstrating the interconnection and simultaneous operation of two types of quantum memory: an atomic-ensemble-based memory and an all-optical loop memory. The former generates and stores single atomic excitations that can then be converted to single photons; and the latter maps incoming photons in and out on demand, at room-temperature and with a broad acceptance bandwidth. Interfacing these two types of quantum memories, we observe a well-preserved quantum cross-correlation, reaching a value of 22, and a violation of the Cauchy-Schwarz inequality up to 549 standard deviations. Furthermore, we demonstrate the creation and storage of a fully operable heralded photon chain state that can achieve memory-built-in combining, swapping, splitting, tuning and chopping single photons in a chain temporally. Such a quantum network allows atomic excitations to be generated, stored, and converted to broadband photons, which are then transferred to the next node, stored, and faithfully retrieved, all at high speed and in a programmable fashion.

quant-ph↗

arXiv2020

Audio-Visual Calibration with Polynomial Regression for 2-D Projection Using SVD-PHAT

This paper proposes a straightforward 2-D method to spatially calibrate the visual field of a camera with the auditory field of an array microphone by generating and overlaying an acoustic image over an optical image. Using a low-cost microphone array and an off-the-shelf camera, we show that polynomial regression can deal efficiently with non-linear camera distortion, and that a recently proposed sound source localization method for real-time processing, SVD-PHAT, can be adapted for this task.

eess.AS↗

arXiv2020

Exocentric to Egocentric Image Generation via Parallel Generative Adversarial Network

Cross-view image generation has been recently proposed to generate images of one view from another dramatically different view. In this paper, we investigate exocentric (third-person) view to egocentric (first-person) view image generation. This is a challenging task since egocentric view sometimes is remarkably different from exocentric view. Thus, transforming the appearances across the two views is a non-trivial task. To this end, we propose a novel Parallel Generative Adversarial Network (P-GAN) with a novel cross-cycle loss to learn the shared information for generating egocentric images from exocentric view. We also incorporate a novel contextual feature loss in the learning procedure to capture the contextual information in images. Extensive experiments on the Exo-Ego datasets show that our model outperforms the state-of-the-art approaches.

cs.CV↗

arXiv2020

Vector Vortex Beam Emitter Embedded in a Photonic Chip

Vector vortex beams simultaneously carrying spin and orbital angular momentum of light promise additional degrees of freedom for modern optics and emerging resources for both classical and quantum information technologies. The inherently infinite dimensions can be exploited to enhance data capacity for sustaining the unprecedented growth in big data and internet traffic, and can be encoded to build quantum computing machines in high-dimensional Hilbert space. So far much progress has been made in the emission of vector vortex beams from a chip surface into free space, however, the generation of vector vortex beams inside a photonic chip hasn't been realized yet. Here, we demonstrate the first vector vortex beam emitter embedded in a photonic chip by using femtosecond laser direct writing. We achieve a conversion of vector vortex beams with an efficiency up to 30% and scalar vortex beams with an efficiency up to 74% from Gaussian beams. We also present an expanded coupled-mode model for understanding the mode conversion and the influence of the imperfection in fabrication. The fashion of embedded generation makes vector vortex beams directly ready for further transmission, manipulation and emission without any additional interconnection. Together with the ability to be integrated as an array, our results may enable vector vortex beams become accessible inside a photonic chip for high-capacity communication and high-dimensional quantum information processing.

quant-ph↗

arXiv2020

AttentionAnatomy: A unified framework for whole-body organs at risk segmentation using multiple partially annotated datasets

Organs-at-risk (OAR) delineation in computed tomography (CT) is an important step in Radiation Therapy (RT) planning. Recently, deep learning based methods for OAR delineation have been proposed and applied in clinical practice for separate regions of the human body (head and neck, thorax, and abdomen). However, there are few researches regarding the end-to-end whole-body OARs delineation because the existing datasets are mostly partially or incompletely annotated for such task. In this paper, our proposed end-to-end convolutional neural network model, called \textbf{AttentionAnatomy}, can be jointly trained with three partially annotated datasets, segmenting OARs from whole body. Our main contributions are: 1) an attention module implicitly guided by body region label to modulate the segmentation branch output; 2) a prediction re-calibration operation, exploiting prior information of the input images, to handle partial-annotation(HPA) problem; 3) a new hybrid loss function combining batch Dice loss and spatially balanced focal loss to alleviate the organ size imbalance problem. Experimental results of our proposed framework presented significant improvements in both Sørensen-Dice coefficient (DSC) and 95\% Hausdorff distance compared to the baseline model.

eess.IV↗

arXiv2020

A Study of Bug Resolution Characteristics in Popular Programming Languages

This paper presents a large-scale study that investigates the bug resolution characteristics among popular Github projects written in different programming languages. We explore correlations but, of course, we cannot infer causation. Specifically, we analyse bug resolution data from approximately 70 million Source Line of Code, drawn from 3 million commits to 600 GitHub projects, primarily written in 10 programming languages. We find notable variations in apparent bug resolution time and patch (fix) size. While interpretation of results from such large-scale empirical studies is inherently difficult, we believe that the differences in medians are sufficiently large to warrant further investigation, replication, re-analysis and follow up research. For example, in our corpus, the median apparent bug resolution time (elapsed time from raise to resolve) for Ruby was 4X that for Go and 2.5X for Java. We also found that patches tend to touch more files for the corpus of strongly typed and for statically typed programs. However, we also found evidence for a lower elapsed resolution time for bug resolution committed to projects constructed from statically typed languages. These findings, if replicated in subsequent follow on studies, may shed further empirical light on the debate about the importance of static typing.

cs.SE↗

arXiv2020

Direct Observation of Quantum Percolation Dynamics

Percolation, describing critical behaviors of phase transition in a geometrical context, prompts wide investigations in natural and social networks as a fundamental model. The introduction of quantum-intrinsic interference and tunneling brings percolation into quantum regime with more fascinating phenomena and unique features, which, however, hasn't been experimentally explored yet. Here we present an experimental demonstration of quantum transport in hexagonal percolation lattices by successfully mapping such large-scale porous structures into a photonic chip using femtosecond laser direct writing techniques. A quantum percolation threshold of 80% is observed in the prototyped laser-written lattices with up to 1,600 waveguides, which is significantly larger than the classical counterpart of 63%. We also investigate the spatial confinement by localization parameters and exhibit the transition from ballistic to diffusive propagation with the decrease of the occupation probability. Direct observation of quantum percolation may deepen the understanding of the relation among materials, quantum transport, geometric quenching, disorder and localization, and inspire applications for quantum technologies.

quant-ph↗

arXiv2019

Improving Semantic Segmentation of Aerial Images Using Patch-based Attention

The trade-off between feature representation power and spatial localization accuracy is crucial for the dense classification/semantic segmentation of aerial images. High-level features extracted from the late layers of a neural network are rich in semantic information, yet have blurred spatial details; low-level features extracted from the early layers of a network contain more pixel-level information, but are isolated and noisy. It is therefore difficult to bridge the gap between high and low-level features due to their difference in terms of physical information content and spatial distribution. In this work, we contribute to solve this problem by enhancing the feature representation in two ways. On the one hand, a patch attention module (PAM) is proposed to enhance the embedding of context information based on a patch-wise calculation of local attention. On the other hand, an attention embedding module (AEM) is proposed to enrich the semantic information of low-level features by embedding local focus from high-level features. Both of the proposed modules are light-weight and can be applied to process the extracted features of convolutional neural networks (CNNs). Experiments show that, by integrating the proposed modules into the baseline Fully Convolutional Network (FCN), the resulting local attention network (LANet) greatly improves the performance over the baseline and outperforms other attention based methods on two aerial image datasets.

cs.CV↗