Diversity Forests: Using Split Sampling to Enable Innovative Complex Split Procedures in Random Forests

The diversity forest algorithm is an alternative candidate node split sampling scheme that makes innovative complex split procedures in random forests possible. While conventional univariable, binary splitting suffices for obtaining strong predictive performance, new complex split procedures can help tackling practically important issues. For example, interactions between features can be exploited effectively by bivariable splitting. With diversity forests, each split is selected from a candidate split set that is sampled in the following way: for $$l = 1, \dots , {nsplits}$$ l = 1 , ⋯ , nsplits : (1) sample one split problem; (2) sample a single or few splits from the split problem sampled in (1) and add this or these splits to the candidate split set. The split problems are specifically structured collections of splits that depend on the respective split procedure considered. This sampling scheme makes innovative complex split procedures computationally tangible while avoiding overfitting. Important general properties of the diversity forest algorithm are evaluated empirically using univariable, binary splitting. Based on 220 data sets with binary outcomes, diversity forests are compared with conventional random forests and random forests using extremely randomized trees. It is seen that the split sampling scheme of diversity forests does not impair the predictive performance of random forests and that the performance is quite robust with regard to the specified nsplits value. The recently developed interaction forests are the first diversity forest method that uses a complex split procedure. Interaction forests allow modeling and detecting interactions between features effectively. Further potential complex split procedures are discussed as an outlook.

Evolution of reproductive strategies in incipient multicellularity

Multicellular organisms can potentially show a large degree of diversity in reproductive strategies, as they could reproduce offspring with varying sizes and compositions compared to their unicellular ancestors. In reality, only a few of these reproductive strategies are prevalent. To understand why this could be the case, we develop a stage-structured population model to probe the evolutionary growth advantages of reproductive strategies in incipient multicellular organisms. The performance of reproductive strategies is evaluated by the growth rates of corresponding populations. We identify the optimal reproductive strategy, which leads to the largest growth rate for a population. Considering the effects of organism size and cellular interaction, we found that distinct reproductive strategies could perform uniquely or equally well under different conditions. Only binary-splitting reproductive strategies can be uniquely optimal. Our results show that organism size and cellular interaction can play crucial roles in shaping reproductive strategies in nascent multicellularity. Our model sheds light on understanding the mechanism driving the evolution of reproductive strategies in incipient multicellularity. Meanwhile, beyond multicellularity, our results imply a crucial factor in the evolution of reproductive strategies of unicellular species - organism size.

Binary and Ternary Fragmentation Analysis of 252Cf Nucleus using Different Nuclear Radii

Pioneering study reveals that a radioactive nucleus may split into two or three fragments and the phenomena are known as binary fission and ternary fission respectively. In order to understand the nuclear stability and related structure aspects, it is of huge interest to explore the fragmentation behavior of a radioactive nucleus in binary and ternary decay modes. In view of this, Binary and ternary fission analysis of 252Cf nucleus is carried out using quantum mechanical fragmentation theory (QMFT). The nuclear potential and Coulomb potential are estimated using different versions of radius vector. The fragmentation structure is found to be independent to the choice of fragment radius for binary as wellas ternary decay paths. The deformation effect is included up to quadrupole (β2) with optimum cold orientations and their influence is explored within binary splitting mode. Moreover, the most probable fission channels explore the role of magic shell effects in binary and ternary fission modes.

A Mixed Method with Effective Color Reduction

This article presents a color quantization technique that combines two previously proposed approaches: the Binary splitting method and the Iterative ant-tree for color quantization method. The resulting algorithm can obtain good quality images with low time consumption. In addition, the iterative nature of the proposed method allows the quality of the quantized image to improve as the iterations progress, although it also allows a good initial image to be quickly obtained. The proposed method was compared to 13 other color quantization techniques and the results showed that it could generate better quantized images than most of the techniques assessed. The statistical significance of the improvement obtained using the new method is confirmed by applying a statistical test to the results of all the methods compared.

The brief comparison of the operational efficiency of pool-testing strategies for COVID-19 mass testing in PCR laboratories

This paper addresses the operational efficiency of different pool-testing strategies in typical scenarios of a PCR laboratory working in mass testing for COVID-19 with different values of prevalence, limitations and conditions of testing, and priorities of optimization. The research employs a model of the laboratory's testing process, created after interviewing of PCR laboratories and studying their operations. The limitations and operational characteristics of this model were applied in a simulation of the testing process with different pool-testing strategies managed by a computer program developed in the LOMT project. The efficiency indicators assessed are the number of assays needed to obtain results of a batch of specimens, the number of specimens identified after the first analysis, and total time to obtain all results. Depending on prevalence, constraints of testing, and priorities of optimization, different pool-testing strategies provide the best operational efficiency. The binary splitting algorithm provides the maximum reduction of the number of assays: from 1.99x reduction for a high prevalence (10%) to 25x reduction for a low prevalence (0.1%), while other algorithms provide the least amount of time to obtain results or the maximum number of the specimens classified after the first analysis.

Analysis and Applications of Adaptive Group Testing Methods for COVID-19

Testing strategies for Covid-19 to maximize number of people tested are urgently needed. Recently, it has been demonstrated that RT-PCR has the sensitivity to detect one positive case in a mixed sample of 32 cases [12], In this paper we propose adaptive group testing strategies based on generalized binary splitting (CBS) [5], where we restrict the group test to the largest group that can be used. The method starts by choosing a group from the population to be tested, performing a test on the combined sample from the entire group, and progressively splitting the group further into subgroups. Compared to individual testing at 4% prevalence, we save 74%; at 1% we save 91%; and at .1% we save 98% of tests. We analyze the number of times each sample is used and show that the method is still efficient if we resort to testing a case individually if the sample is running low.In addition we recommend clinical screening to filter out individuals with symptoms and show this leaves us with a population with lower prevalence. Our approach is particularly applicable to vulnerable confined populations such as nursing homes, prisons, military ships and cruise ships.

Cutting testing costs by the pooling design

Introduction/purpose: The purpose of group testing algorithms is to provide a more rational resource usage. Therefore, it is expected to improve the efficiency of large-scale COVID-19 screening as well. Methods: Two variants of non-adaptive group testing approaches are presented: Hwang's generalized binary-splitting algorithm and the matrix strategy. Results: The positive and negative sides of both approaches are discussed. Also, the estimations of the maximum number of tests are given. The matrix strategy is presented with a particular modification which reduces the corresponding estimation of the maximum number of tests and which does not affect the complexity of the procedure. This modification can be interesting from the applicability viewpoint. Conclusion: Taking into account the current situation, it makes sense to consider these methods in order to achieve some resource cuts in testing, thus making the epidemiological measures more efficient than they are now.