A5100387205- Economic Theory and Institutions
- Insurance and Financial Risk Management
- Complex Systems and Time Series Analysis
- Insurance, Mortality, Demography, Risk Management
- Economic theories and models
- Neural Networks and Applications
- Marine and fisheries research
- HIV/AIDS Impact and Responses
- Neural Networks and Reservoir Computing
- Banking stability, regulation, efficiency
- Economic Theory and Policy
- Energy, Environment, Economic Growth
- Water Quality Monitoring Technologies
- Quantum many-body systems
- Underwater Vehicles and Communication Systems
- Safety and Risk Management
- Advanced Chemical Sensor Technologies
- Global Cancer Incidence and Screening
- Globalization, Economics, and Policies
- Fiscal Policies and Political Economy
- Financial Risk and Volatility Modeling
- Economic Growth and Development
- Multiple and Secondary Primary Cancers
- Quantum Computing Algorithms and Architecture
- Speech and dialogue systems
University of Southern California
2023
Zhejiang University
2014-2022
North China University of Science and Technology
2019-2020
Beijing Information Science & Technology University
2020
Peking University
2020
Peking University Cancer Hospital
2020
The Mountain Institute
2014-2019
University of International Business and Economics
2019
University of North Georgia
2019
Beijing Jiaotong University
2019
Network traffic prediction is a great challenge due to complex statistical properties, generally covering the long-range correlations and self-similarity. To address this issue, article applies an integrated neural computing model predict network traffic, namely, enhanced echo-state restricted Boltzmann machine (eERBM). In structure, possesses following functional components of feature learning, information compensation, input superposition, supervised nonlinear approximation. It motivated...
Abstract Ronald Coase had a profound impact on scholarship worldwide, and not for his ideas alone. Coase's about transaction costs, the nature of firm, role government, problem social cost have been hugely influential. Throughout long life, he also worked to change conduct economics, urging economists ground their conclusions in careful study empirical reality rather than theories that work only blackboard. Less well known, perhaps, is nurture shape emerging fields law economics new...
Support vector machine (SVM) is a powerful learning (ML) technology and the distinctive generalization ability makes it one of most popular approximation tools in field Internet-of-Things (IoT)-based marine data processing. However, SVM has been criticized for trial error parameters, especially, kernel function. How to determine suitable specific problem rather tricky. To give systematic research field, we concentrate on self-adaptive selection functions framework IoT-based prediction....
Marine data prediction plays an increasingly important role in marine environmental monitoring. The support vector machine (SVM) is viewed as a useful learning tool processing, whereas it not completely suitable for the abruptly fluctuating, multi-noise, non-stationary, and abnormal data. To address this issue, paper proposes novel framework sensor prediction, i.e., regression architecture with smoothness priority. This united consistent system functions of acquisition, smoothness, nonlinear...
Abstract This is an introduction to the twelve essays in special memorial issue honor of Ronald Coase. It includes a brief account Coase's long life and some its many achievements. distinctive, worldly empirically-grounded approach economics highlighted, claiming that it has yielded major theoretical policy insights. concludes with summary contributions essays.
Blast furnace gas (BFG) produced from steel industries is generally one of the most important energy supplies in enterprises. Due to a great deal output, fluctuation, and heterogeneity data, it very difficult provide profound insights into its internal dynamic. In this article, novel analysis framework developed for BFG data processing, considering recurrence plot (RP) quantification (RQA). The specific aim investigate relationship between output potential influencing factors. This can be...
Virtual Machines (VMs) and Proof-Carrying Code (PCC) are two techniques that have been used independently to provide safety for (mobile) code. Existing virtual machines, such as the Java VM, several drawbacks: First, effort required verification is considerable. Second more subtly, need by code consumer inhibits amount of optimization can be performed producer. This in turn makes just-in-time compilation surprisingly expensive. Code, on other hand, has its own set limitations, among which...
As a frequent natural disaster, red tide has attracted more and attentions. In fact, results from the joint actions of multiple complex marine environmental factors. Unfortunately, there is no work on interaction analysis between these To inaugurate systematic research this area, novel machine learning based framework developed for series analysis. It combines cross recurrence plot (CRP), quantification (CRQA) statistical This provides general way to transform two into high-dimensional...
As one of the four financial pillars, insurance has functions risk diversification, loss compensation, financing and social management. It is great practical significance to predict level premium income in new normal economy. In this paper, long short-term memory (LSTM) neural network was innovatively applied study prediction. The monthly data China's from January 1999 October 2019 selected for prediction, prediction results were compared with BP network. show that LSTM model can accurately...
Abstract This paper attempts to evaluate the coordinated development state of subsystems within internet financial ecosystem in China from 2011 2016. Focusing on main business modes, technological innovation, and external environment, we select 29 indicators construct an index system adopt a coupling coordination degree model for evaluation. Furthermore, use two weight calculation methods, entropy principal component analysis, ensure robustness results. The empirical results show that...
To evaluate the effects of health insurance status on long-term cancer-specific survival non-small cell lung cancer (NSCLC) in Beijing, China, using a population-based registry data.Information NSCLC patients diagnosed 2008 was derived from Beijing Cancer Registry. The medical records 1,134 cases were sampled and re-surveyed to obtain information potential risk factors. Poorly-insured defined as Uninsured New Rural Cooperative Medical Insurance Scheme (NRCMS), while well-insured included...
Deep learning has become a mainstream method in marine data processing field. However, the raw data, characterized by fluctuations, outliers and noise, is serious obstacle to its performance improvement. To address this problem, hybrid deep computing model developed for sensor prediction, combining smoothing belief echo state network (DBEN). The proposed structure adopts two-stage mode deal with complex characteristics of time series. In preprocessing stage, four methods are considered...
Abstract Quantum many-body simulation provides a straightforward way to understand fundamental physics and connect with quantum information applications. However, suffering from exponentially growing Hilbert space size, characterization in terms of few-body probes real is often insufficient tackle challenging problems such as critical behavior localization (MBL) higher dimensions. Here, we experimentally employ new paradigm on superconducting processor, exploring elusive questions Fock view:...
Purpose This paper aims to revisit the assumption of cyclicality property-liability insurance market and identify a scenario in which so-called underwriting cycles are unpredictable, according dynamic cash flow model generates non-cyclical output dynamics. Design/methodology/approach is on intersection real business cycle models financial cycles. The authors construct an insurer’s flows with stochastic loss shocks capacity constraints, have dual impact both profits access external capital....
We employ impulse response analyses to study aggregated time series data of the US property-casualty insurance market. find that price is more sensitive towards loss shocks than total premiums. Our results at industry level support capacity constraint theory significantly increase and reduce coverage quantity. The firm also examined. a positive relationship between insurers' post-catastrophe performance capital while their ambiguously associated with financial quality losses. findings...
Abstract It is an intimidating challenge, both emotionally and intellectually, to write something of enough worth celebrate the 80th birthday Steven Cheung honor memory Professor Ronald Coase. Both are serious scholars. Coase was a devoted scholar; nothing interested him other than academics. Steve has much wider interests, ranging from photography, calligraphy, art collection. Nonetheless, adamant reformers determined change economics. They have set standard.