arXiv subjects

Chao Yan

Publications and source records attributed to Chao Yan.

At least 73 records · Page 4Linked to original sources

arXiv2024

Optimizing Contrail Detection: A Deep Learning Approach with EfficientNet-b4 Encoding

In the pursuit of environmental sustainability, the aviation industry faces the challenge of minimizing its ecological footprint. Among the key solutions is contrail avoidance, targeting the linear ice-crystal clouds produced by aircraft exhaust. These contrails exacerbate global warming by trapping atmospheric heat, necessitating precise segmentation and comprehensive analysis of contrail images to gauge their environmental impact. However, this segmentation task is complex due to the varying appearances of contrails under different atmospheric conditions and potential misalignment issues in predictive modeling. This paper presents an innovative deep-learning approach utilizing the efficient net-b4 encoder for feature extraction, seamlessly integrating misalignment correction, soft labeling, and pseudo-labeling techniques to enhance the accuracy and efficiency of contrail detection in satellite imagery. The proposed methodology aims to redefine contrail image analysis and contribute to the objectives of sustainable aviation by providing a robust framework for precise contrail detection and analysis in satellite imagery, thus aiding in the mitigation of aviation's environmental impact.

cs.CV↗

arXiv2024

Visualizing orbital angular momentum induced single wavefront dislocation in graphene

Phase singularities are phase-indeterminate points where wave amplitudes are zero, which manifest as phase vertices or wavefront dislocations. In the realm of optical and electron beams, the phase singularity has been extensively explored, demonstrating a profound connection to orbital angular momentum. Direct local imaging of the impact of orbital angular momentum on phase singularities at the nanoscale, however, remains a challenge and has yet to be achieved. Here, we study the role of orbital angular momentum in phase singularities in graphene, particularly at the atomic level, through scanning tunneling microscopy and spectroscopy. Our experiments demonstrate that the scatterings between different orbital angular momentum states, which are induced by local rotational symmetry-breaking potentials, can generate additional phase singularity, and result in robust single wavefront dislocation in real space. Our results pave the way for exploring the effects of orbital degree of freedom on quantum phases in quasiparticle interference processes.

cond-mat.mes-hall↗

arXiv2024

Langmuir turbulence in suspended kelp farms

This study investigates the influence of suspended kelp farms on ocean mixed layer hydrodynamics in the presence of currents and waves. We use the large eddy simulation method, where the wave effect is incorporated by solving the wave-averaged equations. Distinct Langmuir circulation patterns are generated within various suspended farm configurations, including horizontally uniform kelp blocks and spaced kelp rows. Intensified turbulence arises from the farm-generated Langmuir circulation, as opposed to the standard Langmuir turbulence observed without a farm. The creation of Langmuir circulation within the farm is attributed to two primary factors depending on farm configuration: (1) enhanced vertical shear due to kelp frond area density variability, and (2) enhanced lateral shear due to canopy discontinuity at lateral edges of spaced rows. Both enhanced vertical and lateral shear of streamwise velocity, representing the lateral and vertical vorticity components respectively, can be tilted into downstream vorticity to create Langmuir circulation. This vorticity tilting is driven by the Craik- Leibovich vortex force associated with the Stokes drift of surface gravity waves. In addition to the farm-generated Langmuir turbulence, canopy shear layer turbulence is created at the farm bottom edge due to drag discontinuity. The intensity of different types of turbulence depends on both kelp frond area density and the geometric configuration of the farm. The farm-generated turbulence has substantial consequences for nutrient supply and kelp growth. These findings also underscore the significance of the presence of obstacle structures in modifying ocean mixed layer characteristics.

physics.flu-dyn↗

arXiv2023

MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo

Significant strides have been made in enhancing the accuracy of Multi-View Stereo (MVS)-based 3D reconstruction. However, untextured areas with unstable photometric consistency often remain incompletely reconstructed. In this paper, we propose a resilient and effective multi-view stereo approach (MP-MVS). We design a multi-scale windows PatchMatch (mPM) to obtain reliable depth of untextured areas. In contrast with other multi-scale approaches, which is faster and can be easily extended to PatchMatch-based MVS approaches. Subsequently, we improve the existing checkerboard sampling schemes by limiting our sampling to distant regions, which can effectively improve the efficiency of spatial propagation while mitigating outlier generation. Finally, we introduce and improve planar prior assisted PatchMatch of ACMP. Instead of relying on photometric consistency, we utilize geometric consistency information between multi-views to select reliable triangulated vertices. This strategy can obtain a more accurate planar prior model to rectify photometric consistency measurements. Our approach has been tested on the ETH3D High-res multi-view benchmark with several state-of-the-art approaches. The results demonstrate that our approach can reach the state-of-the-art. The associated codes will be accessible at https://github.com/RongxuanTan/MP-MVS.

cs.CV↗

arXiv2023

Split Learning for Distributed Collaborative Training of Deep Learning Models in Health Informatics

Deep learning continues to rapidly evolve and is now demonstrating remarkable potential for numerous medical prediction tasks. However, realizing deep learning models that generalize across healthcare organizations is challenging. This is due, in part, to the inherent siloed nature of these organizations and patient privacy requirements. To address this problem, we illustrate how split learning can enable collaborative training of deep learning models across disparate and privately maintained health datasets, while keeping the original records and model parameters private. We introduce a new privacy-preserving distributed learning framework that offers a higher level of privacy compared to conventional federated learning. We use several biomedical imaging and electronic health record (EHR) datasets to show that deep learning models trained via split learning can achieve highly similar performance to their centralized and federated counterparts while greatly improving computational efficiency and reducing privacy risks.

cs.LG↗

arXiv2023

Private Everlasting Prediction

A private learner is trained on a sample of labeled points and generates a hypothesis that can be used for predicting the labels of newly sampled points while protecting the privacy of the training set [Kasiviswannathan et al., FOCS 2008]. Research uncovered that private learners may need to exhibit significantly higher sample complexity than non-private learners as is the case with, e.g., learning of one-dimensional threshold functions [Bun et al., FOCS 2015, Alon et al., STOC 2019]. We explore prediction as an alternative to learning. Instead of putting forward a hypothesis, a predictor answers a stream of classification queries. Earlier work has considered a private prediction model with just a single classification query [Dwork and Feldman, COLT 2018]. We observe that when answering a stream of queries, a predictor must modify the hypothesis it uses over time, and, furthermore, that it must use the queries for this modification, hence introducing potential privacy risks with respect to the queries themselves. We introduce private everlasting prediction taking into account the privacy of both the training set and the (adaptively chosen) queries made to the predictor. We then present a generic construction of private everlasting predictors in the PAC model. The sample complexity of the initial training sample in our construction is quadratic (up to polylog factors) in the VC dimension of the concept class. Our construction allows prediction for all concept classes with finite VC dimension, and in particular threshold functions with constant size initial training sample, even when considered over infinite domains, whereas it is known that the sample complexity of privately learning threshold functions must grow as a function of the domain size and hence is impossible for infinite domains.

cs.LG↗

arXiv2023

Breakdown of self-cleaning mechanism for nanoscale interfacial substances in tiny-angle twisted bilayer graphene

Realization of high-quality van der Waals (vdW) heterostructures with tailored properties by stacking two-dimensional (2D) layers requires atomically clean interfaces. Because of strong adhesion between the constituent layers, the vdW forces could drive trapped contaminants together into submicron-size bubbles, which leaves large interfacial areas atomically clean. Such a phenomenon is dubbed self-cleaning mechanism in 2D systems. Here, we demonstrate the breakdown of self-cleaning mechanism for nanoscale interfacial bubbles in tiny-angle twisted bilayer graphene (TBG). In the tiny-angle TBG, there is a triangular network of domain boundaries due to structural reconstruction. Our experiments indicate that the bubbles will mainly move along the triangular network of domain boundaries when the sizes of the bubbles are comparable to that of an AA-stacking region in the TBG. When the size of the bubble is smaller than that of an AA-stacking region, the bubble becomes motionless and is fixed in the AA-stacking region because of its large out-of-plane corrugation. Our results reveal a substantial influence of the moiré superlattice on the motion of nanoscale interfacial substances.

cond-mat.mes-hall↗

arXiv2022

Coexistence of Reconstructed and Unreconstructed Structures in Structural Transition Regime of Twisted Bilayer Graphene

In twisted bilayer graphene (TBG), a twist-angle-dependent competition between interlayer stacking energy and intralayer elastic energy results in flat rigid layers at large twist angles and lattice reconstruction at small twist angles. Despite enormous scientific interest and effort in the TBG, however, an experimental study of evolution from the rigid lattice to the reconstructed lattice as a function of twist angle is still missing. Here we present a scanning tunneling microscopy and spectroscopy study to reveal the twist-angle-dependent lattice reconstruction in the TBG. Our experiment demonstrates that there is a transition regime between the rigid regime and the relaxed regime and, unexpectedly, the reconstructed and unreconstructed structures coexist in the transition regime. The coexistence of the two distinct structures in this regime may arise from subtle balance between the interlayer stacking energy and intralayer elastic energy in the TBG with intermediate moiré sizes.

cond-mat.mes-hall↗

arXiv2022

Necessary Conditions in Multi-Server Differential Privacy

We consider protocols where users communicate with multiple servers to perform a computation on the users' data. An adversary exerts semi-honest control over many of the parties but its view is differentially private with respect to honest users. Prior work described protocols that required multiple rounds of interaction or offered privacy against a computationally bounded adversary. Our work presents limitations of non-interactive protocols that offer privacy against unbounded adversaries. We show these protocols demand exponentially more samples for some learning and estimation tasks than centrally private counterparts. This means performing as well as the central model requires interactivity or computational differential privacy, or both.

cs.CR↗

arXiv2022

A Multifaceted Benchmarking of Synthetic Electronic Health Record Generation Models

Synthetic health data have the potential to mitigate privacy concerns when sharing data to support biomedical research and the development of innovative healthcare applications. Modern approaches for data generation based on machine learning, generative adversarial networks (GAN) methods in particular, continue to evolve and demonstrate remarkable potential. Yet there is a lack of a systematic assessment framework to benchmark methods as they emerge and determine which methods are most appropriate for which use cases. In this work, we introduce a generalizable benchmarking framework to appraise key characteristics of synthetic health data with respect to utility and privacy metrics. We apply the framework to evaluate synthetic data generation methods for electronic health records (EHRs) data from two large academic medical centers with respect to several use cases. The results illustrate that there is a utility-privacy tradeoff for sharing synthetic EHR data. The results further indicate that no method is unequivocally the best on all criteria in each use case, which makes it evident why synthetic data generation methods need to be assessed in context.

cs.LG↗

arXiv2022

Distillation to Enhance the Portability of Risk Models Across Institutions with Large Patient Claims Database

Artificial intelligence, and particularly machine learning (ML), is increasingly developed and deployed to support healthcare in a variety of settings. However, clinical decision support (CDS) technologies based on ML need to be portable if they are to be adopted on a broad scale. In this respect, models developed at one institution should be reusable at another. Yet there are numerous examples of portability failure, particularly due to naive application of ML models. Portability failure can lead to suboptimal care and medical errors, which ultimately could prevent the adoption of ML-based CDS in practice. One specific healthcare challenge that could benefit from enhanced portability is the prediction of 30-day readmission risk. Research to date has shown that deep learning models can be effective at modeling such risk. In this work, we investigate the practicality of model portability through a cross-site evaluation of readmission prediction models. To do so, we apply a recurrent neural network, augmented with self-attention and blended with expert features, to build readmission prediction models for two independent large scale claims datasets. We further present a novel transfer learning technique that adapts the well-known method of born-again network (BAN) training. Our experiments show that direct application of ML models trained at one institution and tested at another institution perform worse than models trained and tested at the same institution. We further show that the transfer learning approach based on the BAN produces models that are better than those trained on just a single institution's data. Notably, this improvement is consistent across both sites and occurs after a single retraining, which illustrates the potential for a cheap and general model transfer mechanism of readmission risk prediction.

cs.LG↗

arXiv2022

Molecular Collapse States in Elliptical Graphene/WSe2 Heterostructure Quantum Dots

In relativistic physics, both atomic collapse in heavy nucleus and Hawking radiation in black hole are predicted to occur through Klein tunneling process that couples particles and antiparticles. Recently, atomic collapse states (ACSs) were explicitly realized in graphene because of its relativistic Dirac excitation with large fine structure constant. However, essential role of the Klein tunneling on the ACSs remains elusive in experiment. Here we systematically study the quasibound states in elliptical graphene quantum dots (GQDs). Bonding and antibonding molecular collapse states formed by two coupled ACSs are observed in the elliptical GQDs. Our experiments, supported by theoretical calculations, indicate that the antibonding state of the ACSs will change into a Klein-tunneling-induced quasibound state, revealing deep connection between the ACSs and the Klein tunneling.

cond-mat.mes-hall↗

arXiv2022

Dissolved gas monitoring probe without liquid-gas separation under strong electromagnetic interference

Rapid and direct monitoring of dissolved gases in liquids under strong electromagnetic interference is very important. Electronic gas sensors that can be placed into liquids are difficult to work reliably under strong electromagnetic fields. The existing optical monitoring techniques for dissolved gases all require gas-liquid separation, and are conducted in gas phase, which have poor timeliness. In this paper, a dissolved gas monitoring probe without liquid-gas separation under strong electromagnetic interference is proposed. We take transformer oil-dissolved acetylene monitoring as an example, an oil-core photonic crystal fiber photothermal interferometry probe is proposed and demonstrates the feasibility of trace oil-dissolved acetylene directly monitoring without oil-gas separation. The minimum detection limit reaches 1.4 ppm, and the response time is about 70 minutes. Due to the good insulation performance and the compact size, the all-fiber probe provides applicability to be placed into transformer oil and perform on-line monitoring under strong electromagnetic interference.

physics.ins-det↗

arXiv2022

Seeking Patterns, Not just Memorizing Procedures: Contrastive Learning for Solving Math Word Problems

Math Word Problem (MWP) solving needs to discover the quantitative relationships over natural language narratives. Recent work shows that existing models memorize procedures from context and rely on shallow heuristics to solve MWPs. In this paper, we look at this issue and argue that the cause is a lack of overall understanding of MWP patterns. We first investigate how a neural network understands patterns only from semantics, and observe that, if the prototype equations are the same, most problems get closer representations and those representations apart from them or close to other prototypes tend to produce wrong solutions. Inspired by it, we propose a contrastive learning approach, where the neural network perceives the divergence of patterns. We collect contrastive examples by converting the prototype equation into a tree and seeking similar tree structures. The solving model is trained with an auxiliary objective on the collected examples, resulting in the representations of problems with similar prototypes being pulled closer. We conduct experiments on the Chinese dataset Math23k and the English dataset MathQA. Our method greatly improves the performance in monolingual and multilingual settings.

cs.CL↗

arXiv2022

Dynamically Adjusting Case Reporting Policy to Maximize Privacy and Utility in the Face of a Pandemic

Supporting public health research and the public's situational awareness during a pandemic requires continuous dissemination of infectious disease surveillance data. Legislation, such as the Health Insurance Portability and Accountability Act of 1996 (HIPAA) and recent state-level regulations, permits sharing de-identified person-level data; however, current de-identification approaches are limited. namely, they are inefficient, relying on retrospective disclosure risk assessments, and do not flex with changes in infection rates or population demographics over time. In this paper, we introduce a framework to dynamically adapt de-identification for near-real time sharing of person-level surveillance data. The framework leverages a simulation mechanism, capable of application at any geographic level, to forecast the re-identification risk of sharing the data under a wide range of generalization policies. The estimates inform weekly, prospective policy selection to maintain the proportion of records corresponding to a group size less than 11 (PK11) at or below 0.1. Fixing the policy at the start of each week facilitates timely dataset updates and supports sharing granular date information. We use August 2020 through October 2021 case data from Johns Hopkins University and the Centers for Disease Control and Prevention to demonstrate the framework's effectiveness in maintaining the PK!1 threshold of 0.01. When sharing COVID-19 county-level case data across all US counties, the framework's approach meets the threshold for 96.2% of daily data releases, while a policy based on current de-identification techniques meets the threshold for 32.3%. Periodically adapting the data publication policies preserves privacy while enhancing public health utility through timely updates and sharing epidemiologically critical features.

cs.CR↗

arXiv2022

Adaptive Model Refinement Approach for Bayesian Uncertainty Quantification in Turbulence Model

The Bayesian uncertainty quantification technique has become well established in turbulence modeling over the past few years. However, it is computationally expensive to construct a globally accurate surrogate model for Bayesian inference in a high-dimensional design space, which limits uncertainty quantification for complex flow configurations. Borrowing ideas from stratified sampling and inherited sampling, an adaptive model refinement approach is proposed in this work, which concentrates on asymptotically improving the local accuracy of the surrogate model in the high-posterior-density region by adaptively appending model evaluation points. To achieve this goal, a modification of inherited Latin hypercube sampling is proposed and then integrated into the Bayesian framework. The effectiveness and efficiency of the proposed approach are demonstrated through a two-dimensional heat source inversion problem and its extension to a high-dimensional design space. Compared with the prior-based method, the adaptive model refinement approach has the ability to obtain more reliable inference results using fewer evaluation points. Finally, the approach is applied to parametric uncertainty quantification of the Menter shear-stress transport turbulence model for an axisymmetric transonic bump flow and provides convincing numerical results.

physics.data-an↗

arXiv2021

Pure Differential Privacy from Secure Intermediaries

Recent work in differential privacy has explored the prospect of combining local randomization with a secure intermediary. Specifically, there are a variety of protocols in the secure shuffle model (where an intermediary randomly permutes messages) as well as the secure aggregation model (where an intermediary adds messages). Most of these protocols are limited to approximate differential privacy. An exception is the shuffle protocol by Ghazi, Golowich, Kumar, Manurangsi, Pagh, and Velingker (arXiv:2002.01919): it computes bounded sums under pure differential privacy. Its additive error is $\tilde{O}(1/\varepsilon^{3/2})$, where $\varepsilon$ is the privacy parameter. In this work, we give a new protocol that ensures $O(1/\varepsilon)$ error under pure differential privacy. We also show how to use it to test uniformity of distributions over $[d]$. The tester's sample complexity has an optimal dependence on $d$. Our work relies on a novel class of secure intermediaries which are of independent interest.

cs.CR↗

arXiv2021

Direct Imaging Strain-field Vortex Networks in Twisted Bilayer Graphene Magnified by Moiré Superlattices

In two-dimensional (2D) twisted bilayers, the van der Waals (vdW) interlayer interaction introduces atomic-scale reconstruction at interface by locally rotating lattice to form strain-field vortex networks in their moiré superlattice. However, direct imaging the tiny local lattice rotation of the strain-field vortex requires extremely high spatial resolution and is an outstanding challenge in experiment. Here, a topmost small-period graphene moiré pattern is introduced to magnify sub-Angstrom distortions of the lattice and tiny local lattice rotation in underlying twisted bilayer graphene (TBG). The local periods and low-energy van Hove singularities of the topmost graphene moiré patterns are spatially modified by the atomic-scale reconstruction of the underlying TBG, thus enabling real-space imaging of the strain-field vortex networks. Our results indicate that structure-reconstructed vdW systems can provide a unique substrate to spatially engineer supported two-dimensional materials both in structures and electronic properties.

cond-mat.mes-hall↗