arXiv subjects

Yue Wang

Publications and source records attributed to Yue Wang.

At least 829 records · Page 46Linked to original sources

arXiv2019

An Adaptive Deep Learning Algorithm Based Autoencoder for Interference Channels

Deep learning (DL) based autoencoder has shown great potential to significantly enhance the physical layer performance. In this paper, we present a DL based autoencoder for interference channel. Based on a characterization of a k-user Gaussian interference channel, where the interferences are classified as different levels from weak to very strong interferences based on a coupling parameter α, a DL neural network (NN) based autoencoder is designed to train the data set and decode the received signals. The performance such a DL autoencoder for different interference scenarios are studied, with α known or partially known, where we assume that α is predictable but with a varying up to 10\% at the training stage. The results demonstrate that DL based approach has a significant capability to mitigate the effect induced by a poor signal-to-noise ratio (SNR) and a high interference-to-noise ratio (INR). However, the enhancement depends on the knowledge of α as well as the interference levels. The proposed DL approach performs well with α up to 10\% offset for weak interference level. For strong and very strong interference channel, the offset of α needs to be constrained to less than 5\% and 2\%, respectively, to maintain similar performance as α is known.

cs.LG↗

arXiv2019

E2-Train: Training State-of-the-art CNNs with Over 80% Energy Savings

Convolutional neural networks (CNNs) have been increasingly deployed to edge devices. Hence, many efforts have been made towards efficient CNN inference in resource-constrained platforms. This paper attempts to explore an orthogonal direction: how to conduct more energy-efficient training of CNNs, so as to enable on-device training. We strive to reduce the energy cost during training, by dropping unnecessary computations from three complementary levels: stochastic mini-batch dropping on the data level; selective layer update on the model level; and sign prediction for low-cost, low-precision back-propagation, on the algorithm level. Extensive simulations and ablation studies, with real energy measurements from an FPGA board, confirm the superiority of our proposed strategies and demonstrate remarkable energy savings for training. For example, when training ResNet-74 on CIFAR-10, we achieve aggressive energy savings of >90% and >60%, while incurring a top-1 accuracy loss of only about 2% and 1.2%, respectively. When training ResNet-110 on CIFAR-100, an over 84% training energy saving is achieved without degrading inference accuracy.

cs.LG↗

arXiv2019

Class-Conditional Domain Adaptation on Semantic Segmentation

Semantic segmentation is an important sub-task for many applications, but pixel-level ground truth labeling is costly and there is a tendency to overfit the training data, limiting generalization. Unsupervised domain adaptation can potentially address these problems, allowing systems trained on labelled datasets from one or more source domains (including less expensive synthetic domains) to be adapted to novel target domains. The conventional approach is to automatically align the representational distributions of source and target domains. One limitation of this approach is that it tends to disadvantage lower probability classes. We address this problem by introducing a Class-Conditional Domain Adaptation method (CCDA). It includes a class-conditional multi-scale discriminator and the class-conditional loss. This novel CCDA method encourages the network to shift the domain in a class-conditional manner, and it equalizes loss over classes. We evaluate our CCDA method on two transfer tasks and demonstrate performance comparable to state-of-the-art methods.

cs.CV↗

arXiv2019

DeepGoal: Learning to Drive with driving intention from Human Control Demonstration

Recent research on automotive driving developed an efficient end-to-end learning mode that directly maps visual input to control commands. However, it models distinct driving variations in a single network, which increases learning complexity and is less adaptive for modular integration. In this paper, we re-investigate human's driving style and propose to learn an intermediate driving intention region to relax difficulties in end-to-end approach. The intention region follows both road structure in image and direction towards goal in public route planner, which addresses visual variations only and figures out where to go without conventional precise localization. Then the learned visual intention is projected on vehicle local coordinate and fused with reliable obstacle perception to render a navigation score map widely used for motion planning. The core of the proposed system is a weakly-supervised cGAN-LSTM model trained to learn driving intention from human demonstration. The adversarial loss learns from limited demonstration data with one local planned route and enables reasoning of multi-modal behavior with diverse routes while testing. Comprehensive experiments are conducted with real-world datasets. Results show the proposed paradigm can produce more consistent motion commands with human demonstration, and indicates better reliability and robustness to environment change.

cs.RO↗

arXiv2019

Weakly-Supervised Road Affordances Inference and Learning in Scenes without Traffic Signs

Road attributes understanding is extensively researched to support vehicle's action for autonomous driving, whereas current works mainly focus on urban road nets and rely much on traffic signs. This paper generalizes the same issue to the scenes with little or without traffic signs, such as campuses and residential areas. These scenes face much more individually diverse appearances while few annotated datasets. To explore these challenges, a weakly-supervised framework is proposed to infer and learn road affordances without manual annotation, which includes three attributes of drivable direction, driving attention center and remaining distance. The method consists of two steps: affordances inference from trajectory and learning from partially labeled data. The first step analyzes vehicle trajectories to get partial affordances annotation on image, and the second step implements a weakly-supervised network to learn partial annotation and predict complete road affordances while testing. Real-world datasets are collected to validate the proposed method which achieves 88.2%/80.9% accuracy on direction-level and 74.3% /66.7% accuracy on image-level in familiar and unfamiliar scenes respectively.

cs.RO↗

arXiv2019

Creation of Single Chain of Nanoscale Skyrmion Bubbles with Record-high Temperature Stability in a Geometrically Confined Nanostripe

Nanoscale topologically non-trivial spin textures, such as magnetic skyrmions, have been identified as promising candidates for the transport and storage of information for spintronic applications, notably magnetic racetrack memory devices. The design and realization of single skyrmion chain at room temperature (RT) and above in the low-dimensional nanostructures are of great importance for future practical applications. Here, we report the creation of a single skyrmion bubble chain in a geometrically confined Fe3Sn2 nanostripe with a width comparable to the featured size of a skyrmion bubble. Systematic investigations on the thermal stability have revealed that the single chain of skyrmion bubbles can keep stable at temperatures varying from RT up to a record-high temperature of 630 K. This extreme stability can be ascribed to the weak temperature-dependent magnetic anisotropy and the formation of edge states at the boundaries of the nanostripes. The realization of the highly stable skyrmion bubble chain in a geometrically confined nanostructure is a very important step towards the application of skyrmion-based spintronic devices.

cond-mat.mtrl-sci↗

arXiv2019

Partial Least Squares for Functional Joint Models

Many biomedical studies have identified important imaging biomarkers that are associated with both repeated clinical measures and a survival outcome. The functional joint model (FJM) framework, proposed in Li and Luo (2017), investigates the association between repeated clinical measures and survival data, while adjusting for both high-dimensional images and low-dimensional covariates based upon the functional principal component analysis (FPCA). In this paper, we propose a novel algorithm for the estimation of FJM based on the functional partial least squares (FPLS). Our numerical studies demonstrate that, compared to FPCA, the proposed FPLS algorithm can yield more accurate and robust estimation and prediction performance in many important scenarios. We apply the proposed FPLS algorithm to the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. Data used in the preparation of this article were obtained from the ADNI database.

stat.ME↗

arXiv2019

PointGrow: Autoregressively Learned Point Cloud Generation with Self-Attention

Generating 3D point clouds is challenging yet highly desired. This work presents a novel autoregressive model, PointGrow, which can generate diverse and realistic point cloud samples from scratch or conditioned on semantic contexts. This model operates recurrently, with each point sampled according to a conditional distribution given its previously-generated points, allowing inter-point correlations to be well-exploited and 3D shape generative processes to be better interpreted. Since point cloud object shapes are typically encoded by long-range dependencies, we augment our model with dedicated self-attention modules to capture such relations. Extensive evaluations show that PointGrow achieves satisfying performance on both unconditional and conditional point cloud generation tasks, with respect to realism and diversity. Several important applications, such as unsupervised feature learning and shape arithmetic operations, are also demonstrated.

cs.CV↗

arXiv2019

Embedding Physics Domain Knowledge into a Bayesian Network Enables Layer-by-Layer Process Innovation for Photovoltaics

Process optimization of photovoltaic devices is a time-intensive, trial and error endeavor, without full transparency of the underlying physics, and with user-imposed constraints that may or may not lead to a global optimum. Herein, we demonstrate that embedding physics domain knowledge into a Bayesian network enables an optimization approach that identifies the root cause(s) of underperformance with layer by-layer resolution and reveals alternative optimal process windows beyond global black-box optimization. Our Bayesian-network approach links process conditions to materials descriptors (bulk and interface properties, e.g., bulk lifetime, doping, and surface recombination) and device performance parameters (e.g., cell efficiency), using a Bayesian inference framework with an autoencoder-based surrogate device-physics model that is 100x faster than numerical solvers. With the trained surrogate model, our approach is robust and reduces significantly the time consuming experimentalist intervention, even with small numbers of fabricated samples. To demonstrate our method, we perform layer-by-layer optimization of GaAs solar cells. In a single cycle of learning, we find an improved growth temperature for the GaAs solar cells without any secondary measurements, and demonstrate a 6.5% relative AM1.5G efficiency improvement above baseline and traditional black-box optimization methods.

physics.app-ph↗

arXiv2019

PRNet: Self-Supervised Learning for Partial-to-Partial Registration

We present a simple, flexible, and general framework titled Partial Registration Network (PRNet), for partial-to-partial point cloud registration. Inspired by recently-proposed learning-based methods for registration, we use deep networks to tackle non-convexity of the alignment and partial correspondence problems. While previous learning-based methods assume the entire shape is visible, PRNet is suitable for partial-to-partial registration, outperforming PointNetLK, DCP, and non-learning methods on synthetic data. PRNet is self-supervised, jointly learning an appropriate geometric representation, a keypoint detector that finds points in common between partial views, and keypoint-to-keypoint correspondences. We show PRNet predicts keypoints and correspondences consistently across views and objects. Furthermore, the learned representation is transferable to classification.

cs.LG↗

arXiv2019

Entropic Dynamic Time Warping Kernels for Co-evolving Financial Time Series Analysis

In this work, we develop a novel framework to measure the similarity between dynamic financial networks, i.e., time-varying financial networks. Particularly, we explore whether the proposed similarity measure can be employed to understand the structural evolution of the financial networks with time. For a set of time-varying financial networks with each vertex representing the individual time series of a different stock and each edge between a pair of time series representing the absolute value of their Pearson correlation, our start point is to compute the commute time matrix associated with the weighted adjacency matrix of the network structures, where each element of the matrix can be seen as the enhanced correlation value between pairwise stocks. For each network, we show how the commute time matrix allows us to identify a reliable set of dominant correlated time series as well as an associated dominant probability distribution of the stock belonging to this set. Furthermore, we represent each original network as a discrete dominant Shannon entropy time series computed from the dominant probability distribution. With the dominant entropy time series for each pair of financial networks to hand, we develop a similarity measure based on the classical dynamic time warping framework, for analyzing the financial time-varying networks. We show that the proposed similarity measure is positive definite and thus corresponds to a kernel measure on graphs. The proposed kernel bridges the gap between graph kernels and the classical dynamic time warping framework for multiple financial time series analysis. Experiments on time-varying networks extracted through New York Stock Exchange (NYSE) database demonstrate the effectiveness of the proposed approach.

q-fin.ST↗

arXiv2019

Autonomous Driving using Safe Reinforcement Learning by Incorporating a Regret-based Human Lane-Changing Decision Model

It is expected that many human drivers will still prefer to drive themselves even if the self-driving technologies are ready. Therefore, human-driven vehicles and autonomous vehicles (AVs) will coexist in a mixed traffic for a long time. To enable AVs to safely and efficiently maneuver in this mixed traffic, it is critical that the AVs can understand how humans cope with risks and make driving-related decisions. On the other hand, the driving environment is highly dynamic and ever-changing, and it is thus difficult to enumerate all the scenarios and hard-code the controllers. To face up these challenges, in this work, we incorporate a human decision-making model in reinforcement learning to control AVs for safe and efficient operations. Specifically, we adapt regret theory to describe a human driver's lane-changing behavior, and fit the personalized models to individual drivers for predicting their lane-changing decisions. The predicted decisions are incorporated in the safety constraints for reinforcement learning in training and in implementation. We then use an extended version of double deep Q-network (DDQN) to train our AV controller within the safety set. By doing so, the amount of collisions in training is reduced to zero, while the training accuracy is not impinged.

cs.RO↗

arXiv2019

Computing oscillatory solutions of the Euler system via $\mathcal{K}$-convergence

We develop a method to compute effectively the Young measures associated to sequences of numerical solutions of the compressible Euler system. Our approach is based on the concept of $\mathcal{K}$-convergence adapted to sequences of parametrized measures. The convergence is strong in space and time (a.e.~pointwise or in certain $L^q$ spaces) whereas the measures converge narrowly or in the Wasserstein distance to the corresponding limit.

math.NA↗

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

Intrinsic 2D Ferromagnetism in V5Se8 Epitaxial Thin Films

The discoveries of intrinsic ferromagnetism in atomically-thin van der Waals crystals have opened up a new research field enabling fundamental studies on magnetism at two-dimensional (2D) limit as well as development of magnetic van der Waals heterostructures. To date, a variety of 2D ferromagnetism has been explored mainly by mechanically exfoliating 'originally ferromagnetic (FM)' van der Waals crystals, while bottom-up approach by thin film growth technique has demonstrated emergent 2D ferromagnetism in a variety of 'originally non-FM' van der Waals materials. Here we demonstrate that V5Se8 epitaxial thin films grown by molecular-beam epitaxy (MBE) exhibit emergent 2D ferromagnetism with intrinsic spin polarization of the V 3d electrons despite that the bulk counterpart is 'originally antiferromagnetic (AFM)'. Moreover, thickness-dependence measurements reveal that this newly-developed 2D ferromagnet could be classified as an itinerant 2D Heisenberg ferromagnet with weak magnetic anisotropy, broadening a lineup of 2D magnets to those potentially beneficial for future spintronics applications.

cond-mat.mtrl-sci↗

arXiv2019

Enhancing Model Interpretability and Accuracy for Disease Progression Prediction via Phenotype-Based Patient Similarity Learning

Models have been proposed to extract temporal patterns from longitudinal electronic health records (EHR) for clinical predictive models. However, the common relations among patients (e.g., receiving the same medical treatments) were rarely considered. In this paper, we propose to learn patient similarity features as phenotypes from the aggregated patient-medical service matrix using non-negative matrix factorization. On real-world medical claim data, we show that the learned phenotypes are coherent within each group, and also explanatory and indicative of targeted diseases. We conducted experiments to predict the diagnoses for Chronic Lymphocytic Leukemia (CLL) patients. Results show that the phenotype-based similarity features can improve prediction over multiple baselines, including logistic regression, random forest, convolutional neural network, and more.

stat.AP↗

arXiv2019

Respect Your Emotion: Human-Multi-Robot Teaming based on Regret Decision Model

Often, when modeling human decision-making behaviors in the context of human-robot teaming, the emotion aspect of human is ignored. Nevertheless, the influence of emotion, in some cases, is not only undeniable but beneficial. This work studies the human-like characteristics brought by regret emotion in one-human-multi-robot teaming for the application of domain search. In such application, the task management load is outsourced to the robots to reduce the human's workload, freeing the human to do more important work. The regret decision model is first used by each robot for deciding whether to request human service, then is extended for optimally queuing the requests from multiple robots. For the movement of the robots in the domain search, we designed a path planning algorithm based on dynamic programming for each robot. The simulation shows that the human-like characteristics, namely, risk-seeking and risk-aversion, indeed bring some appealing effects for balancing the workload and performance in the human-multi-robot team.

cs.AI↗

arXiv2019

UGC-VIDEO: perceptual quality assessment of user-generated videos

Recent years have witnessed an ever-expandingvolume of user-generated content (UGC) videos available on the Internet. Nevertheless, progress on perceptual quality assessmentof UGC videos still remains quite limited. There are many distinguished characteristics of UGC videos in the complete video production and delivery chain, and one important property closely relevant to video quality is that there does not exist the pristine source after they are uploaded to the hosting platform,such that they often undergo multiple compression stages before ultimately viewed. To facilitate the UGC video quality assessment,we created a UGC video perceptual quality assessment database. It contains 50 source videos collected from TikTok with diverse content, along with multiple distortion versions generated bythe compression with different quantization levels and coding standards. Subjective quality assessment was conducted to evaluate the video quality. Furthermore, we benchmark the database using existing quality assessment algorithms, and potential roomis observed to future improve the accuracy of UGC video quality measures.

cs.MM↗