Deep neural networks (DNNs) have been applied in class incremental learning, which aims to solve common real-world problems of learning new classes continually. One drawback standard DNNs is that they are prone catastrophic forgetting. Knowledge distillation (KD) a commonly used technique alleviate this problem. In paper, we demonstrate it can indeed help the model output more discriminative results within old classes. However, cannot problem tends classify objects into classes, causing...

Variable annuities are insurance products that contain complex guarantees. To manage the financial risks associated with these guarantees, companies rely heavily on Monte Carlo simulation. However, using simulation to calculate fair market values of guarantees for a large portfolio variable is extremely time consuming. In this article, we propose class GB2 distributions model capture positive skewness typically observed empirically. Numerical results used demonstrate and evaluate performance...

The financial risk associated with the guarantees embedded in variable annuities cannot be addressed adequately by traditional actuarial techniques. Dynamical hedging is used practice to mitigate arising from annuities. However, a major challenge of dynamical calculate dollar Deltas portfolio within short time interval so that rebalancing can done on timely basis. In this article, we propose two-level metamodeling approach efficiently estimating partial under multiasset framework....

Variable annuity is arguably the most complex individual retirement planning product in financial market. Its intricacy stems from a variety of features including investment options, guaranteed benefits, withdrawal etc. In many ways, variable annuities can be viewed as traditional life and products at next level sophistication with added options. Despite significant amount publications by practitioners academics on subject matter, there have been few research papers that systematically...

Data clustering has been discussed extensively, but almost all known conventional algorithms tend to break down in high dimensional spaces because of the inherent sparsity data points. Existing subspace for handling high-dimensional focus on numerical dimensions. In this paper, we designed an iterative algorithm called SUBCAD categorical sets, based minimization objective function clustering. We deduced some cluster memberships changing rules using function. also determine associated with...

Abstract Metamodeling techniques have recently been proposed to address the computational issues related valuation of large portfolios variable annuity contracts. However, it is extremely diffcult, if not impossible, for researchers obtain real datasets frominsurance companies in order test their metamodeling on such and publish results academic journals. To facilitate development dissemination research effcient portfolios, this paper creates a synthetic portfolio contracts based properties...

Abstract Variable annuities contain complex guarantees, whose fair market value cannot be calculated in closed form. To the insurance companies rely heavily on Monte Carlo simulation, which is extremely computationally demanding for large portfolios of variable annuity policies. Metamodeling approaches have been proposed to address these computational issues. An important step metamodeling experimental design that selects a small number representative policies building metamodels. In this...

Data clustering refers to the process of dividing a set objects into homogeneous groups or clusters such that in each cluster are more similar other than those clusters. As one most popular tools for exploratory data analysis, has been applied many scientific areas. In this article, we give review basics clustering, as distance measures and validity, different types algorithms. We also demonstrate applications insurance by using two scalable algorithms, truncated fuzzy c-means (TFCM)...

In the field of biomechanics, delicate structural layout living organisms often brings many inspirations for engineering design. As in case aero-engine piping layout, current single- and multi-tube layouts are ineffective need to be optimised. Inspired by efficient material transfer space utilisation mechanisms biological systems, we propose an automatic pipe method based on co-evolutionary algorithm improved A* algorithm. Taking inspiration from how networks adapt optimize their...

Data clustering is a fundamental machine learning task found in many real-world applications. However, real data usually contain noise or outliers. Handling outliers algorithm can improve the accuracy. In this paper, we propose variant of k-means to provide and outlier detection simultaneously. proposed algorithm, integrated with process achieved via term added objective function algorithm. The generates two partition matrices: one provides cluster groups other be used detect We use both...