arXiv subjects

Jie Min

Publications and source records attributed to Jie Min.

24 records · Page 2Linked to original sources

arXiv2021

Reliability Analysis of Artificial Intelligence Systems Using Recurrent Events Data from Autonomous Vehicles

Artificial intelligence (AI) systems have become increasingly common and the trend will continue. Examples of AI systems include autonomous vehicles (AV), computer vision, natural language processing, and AI medical experts. To allow for safe and effective deployment of AI systems, the reliability of such systems needs to be assessed. Traditionally, reliability assessment is based on reliability test data and the subsequent statistical modeling and analysis. The availability of reliability data for AI systems, however, is limited because such data are typically sensitive and proprietary. The California Department of Motor Vehicles (DMV) oversees and regulates an AV testing program, in which many AV manufacturers are conducting AV road tests. Manufacturers participating in the program are required to report recurrent disengagement events to California DMV. This information is being made available to the public. In this paper, we use recurrent disengagement events as a representation of the reliability of the AI system in AV, and propose a statistical framework for modeling and analyzing the recurrent events data from AV driving tests. We use traditional parametric models in software reliability and propose a new nonparametric model based on monotonic splines to describe the event process. We develop inference procedures for selecting the best models, quantifying uncertainty, and testing heterogeneity in the event process. We then analyze the recurrent events data from four AV manufacturers, and make inferences on the reliability of the AI systems in AV. We also describe how the proposed analysis can be applied to assess the reliability of other AI systems.

cs.AI↗

arXiv2021

Local geometry of symplectic divisors with applications to contact torus bundles

In this note we study the contact geometry of symplectic divisors. We show the contact structure induced on the boundary of a divisor neighborhood is invariant under toric and interior blow-ups and blow-downs. We also construct an open book decomposition on the boundary of a concave divisor neighborhood and apply it to the study of universally tight contact structures of contact torus bundles.

math.SG↗

arXiv2020

Refined Tail Asymptotic Properties for the $M^X/G/1$ Retrial Queue

In the literature, retrial queues with batch arrivals and heavy service times have been studied and the so-called equivalence theorem has been established under the condition that the service time is heavier than the batch size. The equivalence theorem provides the distribution (or tail) equivalence between the total number of customers in the system for the retrial queue and the total number of customers in the corresponding standard (non-retrial) queue. In this paper, under the assumption of regularly varying tails, we eliminate this condition by allowing that the service time can be either heavier or lighter than the batch size. The main contribution made in this paper is an asymptotic characterization of the difference between two tail probabilities: the probability of the total number of customers in the system for the $M^X/G/1$ retrial queue and the probability of the total number of customers in the corresponding standard (non-retrial) queue. The equivalence theorem by allowing a heavier batch size is another contribution in this paper.

math.PR↗

arXiv2020

Nested Scale Editing for Conditional Image Synthesis

We propose an image synthesis approach that provides stratified navigation in the latent code space. With a tiny amount of partial or very low-resolution image, our approach can consistently out-perform state-of-the-art counterparts in terms of generating the closest sampled image to the ground truth. We achieve this through scale-independent editing while expanding scale-specific diversity. Scale-independence is achieved with a nested scale disentanglement loss. Scale-specific diversity is created by incorporating a progressive diversification constraint. We introduce semantic persistency across the scales by sharing common latent codes. Together they provide better control of the image synthesis process. We evaluate the effectiveness of our proposed approach through various tasks, including image outpainting, image superresolution, and cross-domain image translation.

cs.CV↗

arXiv2019

Prediction of large barocaloric effects in thermoelectric superionic materials

We predict the existence of large barocaloric effects above room temperature in the thermoelectric fast-ion conductor Cu$_{2}$Se by using classical molecular dynamics simulations and first-principles computational methods. A hydrostatic pressure of $1$ GPa induces large isothermal entropy changes of $|ΔS| \sim 15$-$45$ Jkg$^{-1}$K$^{-1}$ and adiabatic temperature shifts of $|ΔT| \sim 10$ K in the temperature interval $400 \le T \le 700$ K. Structural phase transitions are absent in the analysed thermodynamic range. The causes of such large barocaloric effects are significant $P$-induced variations on the ionic conductivity of Cu$_{2}$Se and the inherently high anharmonicity of the material. Uniaxial stresses of the same magnitude, either compressive or tensile, produce comparatively much smaller caloric effects, namely, $|ΔS| \sim 1$ Jkg$^{-1}$K$^{-1}$ and $|ΔT| \sim 0.1$ K, due to practically null influence on the ionic diffusivity of Cu$_{2}$Se. Our simulation work shows that thermoelectric compounds presenting high ionic disorder, like copper and silver-based chalcogenides, may render large mechanocaloric effects and thus are promising materials for engineering solid-state cooling applications that do not require the application of electric fields.

cond-mat.mtrl-sci↗

arXiv2019

Liquid Warping GAN: A Unified Framework for Human Motion Imitation, Appearance Transfer and Novel View Synthesis

We tackle the human motion imitation, appearance transfer, and novel view synthesis within a unified framework, which means that the model once being trained can be used to handle all these tasks. The existing task-specific methods mainly use 2D keypoints (pose) to estimate the human body structure. However, they only expresses the position information with no abilities to characterize the personalized shape of the individual person and model the limbs rotations. In this paper, we propose to use a 3D body mesh recovery module to disentangle the pose and shape, which can not only model the joint location and rotation but also characterize the personalized body shape. To preserve the source information, such as texture, style, color, and face identity, we propose a Liquid Warping GAN with Liquid Warping Block (LWB) that propagates the source information in both image and feature spaces, and synthesizes an image with respect to the reference. Specifically, the source features are extracted by a denoising convolutional auto-encoder for characterizing the source identity well. Furthermore, our proposed method is able to support a more flexible warping from multiple sources. In addition, we build a new dataset, namely Impersonator (iPER) dataset, for the evaluation of human motion imitation, appearance transfer, and novel view synthesis. Extensive experiments demonstrate the effectiveness of our method in several aspects, such as robustness in occlusion case and preserving face identity, shape consistency and clothes details. All codes and datasets are available on https://svip-lab.github.io/project/impersonator.html

cs.CV↗