Searcharxiv⌕ Search

arXiv subjects

Yue Wang

Publications and source records attributed to Yue Wang.

At least 811 records · Page 45Linked to original sources
arXiv2020

Learning hierarchical behavior and motion planning for autonomous driving

Learning-based driving solution, a new branch for autonomous driving, is expected to simplify the modeling of driving by learning the underlying mechanisms from data. To improve the tactical decision-making for learning-based driving solution, we introduce hierarchical behavior and motion planning (HBMP) to explicitly model the behavior in learning-based solution. Due to the coupled action space of behavior and motion, it is challenging to solve HBMP problem using reinforcement learning (RL) for long-horizon driving tasks. We transform HBMP problem by integrating a classical sampling-based motion planner, of which the optimal cost is regarded as the rewards for high-level behavior learning. As a result, this formulation reduces action space and diversifies the rewards without losing the optimality of HBMP. In addition, we propose a sharable representation for input sensory data across simulation platforms and real-world environment, so that models trained in a fast event-based simulator, SUMO, can be used to initialize and accelerate the RL training in a dynamics based simulator, CARLA. Experimental results demonstrate the effectiveness of the method. Besides, the model is successfully transferred to the real-world, validating the generalization capability.

cs.RO↗
arXiv2020

Generative Temporal Link Prediction via Self-tokenized Sequence Modeling

We formalize networks with evolving structures as temporal networks and propose a generative link prediction model, Generative Link Sequence Modeling (GLSM), to predict future links for temporal networks. GLSM captures the temporal link formation patterns from the observed links with a sequence modeling framework and has the ability to generate the emerging links by inferring from the probability distribution on the potential future links. To avoid overfitting caused by treating each link as a unique token, we propose a self-tokenization mechanism to transform each raw link in the network to an abstract aggregation token automatically. The self-tokenization is seamlessly integrated into the sequence modeling framework, which allows the proposed GLSM model to have the generalization capability to discover link formation patterns beyond raw link sequences. We compare GLSM with the existing state-of-art methods on five real-world datasets. The experimental results demonstrate that GLSM obtains future positive links effectively in a generative fashion while achieving the best performance (2-10\% improvements on AUC) among other alternatives.

cs.LG↗
arXiv2020

Direct Speech-to-image Translation

Direct speech-to-image translation without text is an interesting and useful topic due to the potential applications in human-computer interaction, art creation, computer-aided design. etc. Not to mention that many languages have no writing form. However, as far as we know, it has not been well-studied how to translate the speech signals into images directly and how well they can be translated. In this paper, we attempt to translate the speech signals into the image signals without the transcription stage. Specifically, a speech encoder is designed to represent the input speech signals as an embedding feature, and it is trained with a pretrained image encoder using teacher-student learning to obtain better generalization ability on new classes. Subsequently, a stacked generative adversarial network is used to synthesize high-quality images conditioned on the embedding feature. Experimental results on both synthesized and real data show that our proposed method is effective to translate the raw speech signals into images without the middle text representation. Ablation study gives more insights about our method.

cs.MM↗
arXiv2020

Universal Adversarial Perturbations Generative Network for Speaker Recognition

Attacking deep learning based biometric systems has drawn more and more attention with the wide deployment of fingerprint/face/speaker recognition systems, given the fact that the neural networks are vulnerable to the adversarial examples, which have been intentionally perturbed to remain almost imperceptible for human. In this paper, we demonstrated the existence of the universal adversarial perturbations~(UAPs) for the speaker recognition systems. We proposed a generative network to learn the mapping from the low-dimensional normal distribution to the UAPs subspace, then synthesize the UAPs to perturbe any input signals to spoof the well-trained speaker recognition model with high probability. Experimental results on TIMIT and LibriSpeech datasets demonstrate the effectiveness of our model.

eess.AS↗
arXiv2020

Learning to fool the speaker recognition

Due to the widespread deployment of fingerprint/face/speaker recognition systems, attacking deep learning based biometric systems has drawn more and more attention. Previous research mainly studied the attack to the vision-based system, such as fingerprint and face recognition. While the attack for speaker recognition has not been investigated yet, although it has been widely used in our daily life. In this paper, we attempt to fool the state-of-the-art speaker recognition model and present \textit{speaker recognition attacker}, a lightweight model to fool the deep speaker recognition model by adding imperceptible perturbations onto the raw speech waveform. We find that the speaker recognition system is also vulnerable to the attack, and we achieve a high success rate on the non-targeted attack. Besides, we also present an effective method to optimize the speaker recognition attacker to obtain a trade-off between the attack success rate with the perceptual quality. Experiments on the TIMIT dataset show that we can achieve a sentence error rate of $99.2\%$ with an average SNR $57.2\text{dB}$ and PESQ 4.2 with speed rather faster than real-time.

eess.AS↗
arXiv2020

Globally optimal consensus maximization for robust visual inertial localization in point and line map

Map based visual inertial localization is a crucial step to reduce the drift in state estimation of mobile robots. The underlying problem for localization is to estimate the pose from a set of 3D-2D feature correspondences, of which the main challenge is the presence of outliers, especially in changing environment. In this paper, we propose a robust solution based on efficient global optimization of the consensus maximization problem, which is insensitive to high percentage of outliers. We first introduce translation invariant measurements (TIMs) for both points and lines to decouple the consensus maximization problem into rotation and translation subproblems, allowing for a two-stage solver with reduced solution dimensions. Then we show that (i) the rotation can be calculated by minimizing TIMs using only 1-dimensional branch-and-bound (BnB), (ii) the translation can be found by running 1-dimensional search for three times with prioritized progressive voting. Compared with the popular randomized solver, our solver achieves deterministic global convergence without depending on an initial value. While compared with existing BnB based methods, ours is exponentially faster. Finally, by evaluating the performance on both simulation and real-world datasets, our approach gives accurate pose even when there are 90\% outliers (only 2 inliers).

cs.RO↗
arXiv2020

When Relation Networks meet GANs: Relation GANs with Triplet Loss

Though recent research has achieved remarkable progress in generating realistic images with generative adversarial networks (GANs), the lack of training stability is still a lingering concern of most GANs, especially on high-resolution inputs and complex datasets. Since the randomly generated distribution can hardly overlap with the real distribution, training GANs often suffers from the gradient vanishing problem. A number of approaches have been proposed to address this issue by constraining the discriminator's capabilities using empirical techniques, like weight clipping, gradient penalty, spectral normalization etc. In this paper, we provide a more principled approach as an alternative solution to this issue. Instead of training the discriminator to distinguish real and fake input samples, we investigate the relationship between paired samples by training the discriminator to separate paired samples from the same distribution and those from different distributions. To this end, we explore a relation network architecture for the discriminator and design a triplet loss which performs better generalization and stability. Extensive experiments on benchmark datasets show that the proposed relation discriminator and new loss can provide significant improvement on variable vision tasks including unconditional and conditional image generation and image translation.

cs.CV↗
arXiv2020

Effect of magnetic field on transports of charged particles in the weakly ionized plasma with power-law q-distributions in nonextensive statistics

By using the generalized Boltzmann equation of transport in nonextensive statistics, we study transport properties of the diffusion flux and the heat flux of charged particles in the weakly ionized plasma with the power-law q-distributions under the magnetic field. We derive the tensor expressions of diffusion coefficient, thermal diffusion coefficient, mobility and thermal conductivity of electrons and ions in the q-distributed plasma under magnetic field. We show that the tensors of the diffusion coefficient, the thermal diffusion coefficient and the thermal conductivity are strongly depend on the q-parameters in nonextensive statistics, and so they are generally not the same as those in the magnetized plasma with a Maxwell distribution.

physics.plasm-ph↗
arXiv2020

When does the onset of multiple stellar populations in star clusters occur-III: No evidence of significant chemical variations in main-sequence stars of NGC 419

Recent studies have revealed that the onset age for the presence of multiple stellar populations (MPs) in star clusters seems to correspond to the disappearance of the extended main-sequence turnoff (eMSTO) in young clusters, a pattern associated with stellar rotations. A speculative suggestion is that MPs might be caused by the magnetic brake, a stellar evolutionary effect linked to the rotation. In this work, we use the young massive cluster NGC 419 as a testbed. We examined if its magnetically baked MS stars would exhibit MPs. Using the deep ultraviolet and visible images observed through the Hubble Space Telescope, combined with a specific color index that is sensitive to the nitrogen (N) abundance, we examined if its late G- and K-type MS stars are affected by N variation. Our analysis reports that the morphology of its GK-type MS is most likely an SSP, and only a negligible probability that indicates a N variation up to 0.4 dex is present. We conclude that there is no significant N variation among its GK-type MS stars. The absence of a significant chemical variation among the late-type MS stars indicates that MPs might not be a specific pattern of magnetically braked stars.

astro-ph.SR↗
arXiv2020

Causal inference in degenerate systems: An impossibility result

Causal relationships among variables are commonly represented via directed acyclic graphs. There are many methods in the literature to quantify the strength of arrows in a causal acyclic graph. These methods, however, have undesirable properties when the causal system represented by a directed acyclic graph is degenerate. In this paper, we characterize a degenerate causal system using multiplicity of Markov boundaries. We show that in this case, it is impossible to find an identifiable quantitative measure of causal effects that satisfy a set of natural criteria. To supplement the impossibility result, we also develop algorithms to identify degenerate causal systems from observed data. Performance of our algorithms is investigated through synthetic data analysis.

math.ST↗
arXiv2020

2-Entity RANSAC for robust visual localization in changing environment

Visual localization has attracted considerable attention due to its low-cost and stable sensor, which is desired in many applications, such as autonomous driving, inspection robots and unmanned aerial vehicles. However, current visual localization methods still struggle with environmental changes across weathers and seasons, as there is significant appearance variation between the map and the query image. The crucial challenge in this situation is that the percentage of outliers, i.e. incorrect feature matches, is high. In this paper, we derive minimal closed form solutions for 3D-2D localization with the aid of inertial measurements, using only 2 pairs of point matches or 1 pair of point match and 1 pair of line match. These solutions are further utilized in the proposed 2-entity RANSAC, which is more robust to outliers as both line and point features can be used simultaneously and the number of matches required for pose calculation is reduced. Furthermore, we introduce three feature sampling strategies with different advantages, enabling an automatic selection mechanism. With the mechanism, our 2-entity RANSAC can be adaptive to the environments with different distribution of feature types in different segments. Finally, we evaluate the method on both synthetic and real-world datasets, validating its performance and effectiveness in inter-session scenarios.

cs.RO↗
arXiv2020

Blow-up criterion for the 2-D Prandtl equation

In this paper, we consider the 2-D Prandtl equation with constant outer flow and monotonic data. We prove that if the curvature of the velocity distribution(i.e., $\partial_y^2u$) is bounded near the boundary, then the solution can not develop the singularity.

math.AP↗
arXiv2020

Cellular Decomposition for Non-repetitive Coverage Task with Minimum Discontinuities

A mechanism to derive non-repetitive coverage path solutions with a proven minimal number of discontinuities is proposed in this work, with the aim to avoid unnecessary, costly end effector lift-offs for manipulators. The problem is motivated by the automatic polishing of an object. Due to the non-bijective mapping between the workspace and the joint-space, a continuous coverage path in the workspace may easily be truncated in the joint-space, incuring undesirable end effector lift-offs. Inversely, there may be multiple configuration choices to cover the same point of a coverage path through the solution of the Inverse Kinematics. The solution departs from the conventional local optimisation of the coverage path shape in task space, or choosing appropriate but possibly disconnected configurations, to instead explicitly explore the leaast number of discontinuous motions through the analysis of the structure of valid configurations in joint-space. The two novel contributions of this paper include proof that the least number of path discontinuities is predicated on the surrounding environment, independent from the choice of the actual coverage path; thus has a minimum. And an efficient finite cellular decomposition method to optimally divide the workspace into the minimum number of cells, each traversable without discontinuties by any arbitrary coverage path within. Extensive simulation examples and real-world results on a 5 DoF manipulator are presented to prove the validity of the proposed strategy in realistic settings.

cs.RO↗
arXiv2020

DAF-NET: a saliency based weakly supervised method of dual attention fusion for fine-grained image classification

Fine-grained image classification is a challenging problem, since the difficulty of finding discriminative features. To handle this circumstance, basically, there are two ways to go. One is use attention based method to focus on informative areas, while the other one aims to find high order between features. Further, for attention based method there are two directions, activation based and detection based, which are proved effective by scholars. However ,rare work focus on fusing two types of attention with high order feature. In this paper, we propose a novel DAF method which fuse two types of attention and use them to as PAF(part attention filter) in deep bilinear transformation module to mine the relationship between separate parts of an object. Briefly, our network constructed by a student net who attempt to output two attention maps and a teacher net uses these two maps as empirical information to refine the result. The experiment result shows that only student net could get 87.6% accuracy in CUB dataset while cooperating with teacher net could achieve 89.1% accuracy.

cs.CV↗
arXiv2020

Fractional Skipping: Towards Finer-Grained Dynamic CNN Inference

While increasingly deep networks are still in general desired for achieving state-of-the-art performance, for many specific inputs a simpler network might already suffice. Existing works exploited this observation by learning to skip convolutional layers in an input-dependent manner. However, we argue their binary decision scheme, i.e., either fully executing or completely bypassing one layer for a specific input, can be enhanced by introducing finer-grained, "softer" decisions. We therefore propose a Dynamic Fractional Skipping (DFS) framework. The core idea of DFS is to hypothesize layer-wise quantization (to different bitwidths) as intermediate "soft" choices to be made between fully utilizing and skipping a layer. For each input, DFS dynamically assigns a bitwidth to both weights and activations of each layer, where fully executing and skipping could be viewed as two "extremes" (i.e., full bitwidth and zero bitwidth). In this way, DFS can "fractionally" exploit a layer's expressive power during input-adaptive inference, enabling finer-grained accuracy-computational cost trade-offs. It presents a unified view to link input-adaptive layer skipping and input-adaptive hybrid quantization. Extensive experimental results demonstrate the superior tradeoff between computational cost and model expressive power (accuracy) achieved by DFS. More visualizations also indicate a smooth and consistent transition in the DFS behaviors, especially the learned choices between layer skipping and different quantizations when the total computational budgets vary, validating our hypothesis that layer quantization could be viewed as intermediate variants of layer skipping. Our source code and supplementary material are available at \link{https://github.com/Torment123/DFS}.

cs.LG↗
arXiv2020

Dual Dynamic Inference: Enabling More Efficient, Adaptive and Controllable Deep Inference

State-of-the-art convolutional neural networks (CNNs) yield record-breaking predictive performance, yet at the cost of high-energy-consumption inference, that prohibits their widely deployments in resource-constrained Internet of Things (IoT) applications. We propose a dual dynamic inference (DDI) framework that highlights the following aspects: 1) we integrate both input-dependent and resource-dependent dynamic inference mechanisms under a unified framework in order to fit the varying IoT resource requirements in practice. DDI is able to both constantly suppress unnecessary costs for easy samples, and to halt inference for all samples to meet hard resource constraints enforced; 2) we propose a flexible multi-grained learning to skip (MGL2S) approach for input-dependent inference which allows simultaneous layer-wise and channel-wise skipping; 3) we extend DDI to complex CNN backbones such as DenseNet and show that DDI can be applied towards optimizing any specific resource goals including inference latency or energy cost. Extensive experiments demonstrate the superior inference accuracy-resource trade-off achieved by DDI, as well as the flexibility to control such trade-offs compared to existing peer methods. Specifically, DDI can achieve up to 4 times computational savings with the same or even higher accuracy as compared to existing competitive baselines.

cs.LG↗
arXiv2019

Reducing Selection Bias in Counterfactual Reasoning for Individual Treatment Effects Estimation

Counterfactual reasoning is an important paradigm applicable in many fields, such as healthcare, economics, and education. In this work, we propose a novel method to address the issue of \textit{selection bias}. We learn two groups of latent random variables, where one group corresponds to variables that only cause selection bias, and the other group is relevant for outcome prediction. They are learned by an auto-encoder where an additional regularized loss based on Pearson Correlation Coefficient (PCC) encourages the de-correlation between the two groups of random variables. This allows for explicitly alleviating selection bias by only keeping the latent variables that are relevant for estimating individual treatment effects. Experimental results on a synthetic toy dataset and a benchmark dataset show that our algorithm is able to achieve state-of-the-art performance and improve the result of its counterpart that does not explicitly model the selection bias.

cs.LG↗