arXiv subjects

Yubo Wang

Publications and source records attributed to Yubo Wang.

99 records · Page 6Linked to original sources

arXiv2021

Detecting Mitosis against Domain Shift using a Fused Detector and Deep Ensemble Classification Model for MIDOG Challenge

Mitotic figure count is an important marker of tumor proliferation and has been shown to be associated with patients' prognosis. Deep learning based mitotic figure detection methods have been utilized to automatically locate the cell in mitosis using hematoxylin \& eosin (H\&E) stained images. However, the model performance deteriorates due to the large variation of color tone and intensity in H\&E images. In this work, we proposed a two stage mitotic figure detection framework by fusing a detector and a deep ensemble classification model. To alleviate the impact of color variation in H\&E images, we utilize both stain normalization and data augmentation, aiding model to learn color irrelevant features. The proposed model obtains an F1 score of 0.7550 on the preliminary testing set released by the MIDOG challenge.

cs.CV↗

arXiv2021

Efficient and tunable blue light generation using lithium niobate nonlinear photonics

Thin film lithium niobate (LN) has recently emerged as a playground for chip-scale nonlinear optics and leads to highly efficient frequency conversions from near infrared to near-visible bands. For many nonlinear and quantum photonics applications, it is desirable to operate deep into the visible band within LN's transparency window. However, the strong material dispersion at short wavelengths makes phase-matching difficult, necessitating sub-micron scale control of domain structures for efficient quasi-phase-matching (QPM). Here we report the operation of thin film LN in the blue wavelength and high fidelity poling of thin-film LN waveguide to this regime. As a result, quasi-phase matching is realized between IR (871nm) and blue (435.5nm) wavelengths in a straight waveguide and prompts strong blue light generation with a conversion efficiency $2900\pm400\%W^{-1}cm^{-2}$

physics.optics↗

arXiv2020

An asymptotic-preserving dynamical low-rank method for the multi-scale multi-dimensional linear transport equation

We introduce a dynamical low-rank method to reduce the computational complexity for solving the multi-scale multi-dimensional linear transport equation. The method is based on a macro-micro decomposition of the equation. The proposed numerical method uses the low rank approximation only for the micro part of the solution. The time and spatial discretizations are done properly so that the overall scheme is second order accurate and asymptotic-preserving (AP); that is, in the diffusive regime, the scheme becomes a macroscopic solver for the limiting diffusion equation and is automatically low rank. We demonstrate the accuracy and efficiency of the proposed low rank method by a number of two-dimensional examples.

math.NA↗

arXiv2020

Uncertainty quantification for the BGK model of the Boltzmann equation using multilevel variance reduced Monte Carlo methods

We propose a control variate multilevel Monte Carlo method for the kinetic BGK model of the Boltzmann equation subject to random inputs. The method combines a multilevel Monte Carlo technique with the computation of the optimal control variate multipliers derived from local or global variance minimization problems. Consistency and convergence analysis for the method equipped with a second-order positivity-preserving and asymptotic-preserving scheme in space and time is also performed. Various numerical examples confirm that the optimized multilevel Monte Carlo method outperforms the classical multilevel Monte Carlo method especially for problems with discontinuities.

math.NA↗

arXiv2018

Battery Life-Cycle Optimization and Runtime Control for Commercial Buildings Demand Side Management: A New York City Case Study

In metropolitan areas populated with commercial buildings, electric power supply is stringent especially during business hours. Demand side management using battery is a promising solution to mitigate peak demands, however long payback time creates barriers for large scale adoption. In this paper, we have developed a design phase battery life-cycle cost assessment tool and a runtime controller for the building owners, taking into account the degradation of battery. In the design phase, perfect knowledge on building load profile is assumed to estimate ideal payback time. In runtime, stochastic programming and load predictions are applied to address the uncertainties in loads for producing optimal battery operation. For validation, we have performed numerical experiments using the real-life tariff model serves New York City, Zn/MnO2 battery, and state-of-the-art building simulation tool. Experimental results shows a small gap between design phase assessment and runtime control. To further examine the proposed methods, we have applied the same tariff model and performed numerical experiments on nine weather zones and three types of commercial buildings. On contrary to the common practice of shallow discharging battery for preventing phenomenal degradation, experimental results show promising payback time achieved by optimally deep discharge a battery.

eess.SP↗

arXiv2016

Optimal Operation of Stationary and Mobile Batteries in Distribution Grids

The trending integrations of Battery Energy Storage System (BESS, stationary battery) and Electric Vehicles (EV, mobile battery) to distribution grids call for advanced Demand Side Management (DSM) technique that addresses the scalability concerns of the system and stochastic availabilities of EVs. Towards this goal, a stochastic DSM is proposed to capture the uncertainties in EVs. Numerical approximation is then used to make the problem tractable. To accelerate the computational speed, the proposed DSM is tightly relaxed to a convex form using second-order cone programming. Furthermore, in light of the continuous increasing problem size, a distributed method with a guaranteed convergence is applied to shift the centralized computational burden to distributed controllers. To verify the proposed DSM, real-life EV data collected on UCLA campus is used to test the proposed DSM in an IEEE benchmark test system. Numerical results demonstrate the correctness and merits of the proposed approach.

eess.SY↗

arXiv2015

Two-Tier Prediction of Solar Power Generation with Limited Sensing Resource

This paper considers a typical solar installations scenario with limited sensing resources. In the literature, there exist either day-ahead solar generation prediction methods with limited accuracy, or high accuracy short timescale methods that are not suitable for applications requiring longer term prediction. We propose a two-tier (global-tier and local-tier) prediction method to improve accuracy for long term (24 hour) solar generation prediction using only the historical power data. In global-tier, we examine two popular heuristic methods: weighted k-Nearest Neighbors (k-NN) and Neural Network (NN). In local-tier, the global-tier results are adaptively updated using real-time analytical residual analysis. The proposed method is validated using the UCLA Microgrid with 35kW of solar generation capacity. Experimental results show that the proposed two-tier prediction method achieves higher accuracy compared to day-ahead predictions while providing the same prediction length. The difference in the overall prediction performance using either weighted k-NN based or NN based in the global-tier are carefully discussed and reasoned. Case studies with a typical sunny day and a cloudy day are carried out to demonstrate the effectiveness of the proposed two-tier predictions.

stat.AP↗

arXiv2011

Dynamics of competing ideas in complex social systems

Individuals accepting an idea may intentionally or unintentionally impose influences in a certain neighborhood area, making other individuals within the area less likely or even impossible to accept other competing ideas. Depending on whether such influences strictly prohibit neighborhood individuals from accepting other ideas or not, we classify them into exclusive and non-exclusive influences, respectively. Our study reveals for the first time the rich and complex dynamics of two competing ideas with neighborhood influences in scale-free social networks: depending on whether they have exclusive or non-exclusive influences, the final state varies from multiple coexistence to founder control to exclusion, with different sizes of population accepting each of the ideas respectively. Such results provide insights helpful for better understanding the spread (and the control of spread) of ideas in human society.

physics.soc-ph↗

arXiv2011

Effects of fear factors in disease propagation

Upon an outbreak of a dangerous infectious disease, people generally tend to reduce their contacts with others in fear of getting infected. Such typical actions apparently help slow down the spreading of infection. Thanks to today's broad public media coverage, the fear factor may even contribute to prevent an outbreak from happening. We are motivated to study such effects by adopting a complex network approach. Firstly we evaluate the simple case where connections between individuals are randomly removed due to fear factor. Then we consider a different case where each individual keeps at least a few connections after contact reduction. Such a case is arguably more realistic since people may choose to keep a few social contacts, e.g., with their family members and closest friends, at any cost. Finally a study is conducted on the case where connection removals are carried out dynamically while the infection is spreading out. Analytical and simulation results show that the fear factor may not easily prevent an epidemic outbreak from happening in scale-free networks. However, it significantly reduces the fraction of the nodes ever getting infected during the outbreak.

physics.soc-ph↗