Deep neural nets with a large number of parameters are very powerful machine learning systems. However, overfitting is a serious problem in such networks. Large networks are also slow to use, making it difficult to deal with overfitting by combining the predictions of many different large neural nets at test time. Dropout is a technique for addressing this problem. The key idea is to randomly drop units (along with their connections) from the neural network during training. This prevents units from co-adapting too much. During training, dropout samples from an exponential number of different "thinned" networks. At test time, it is easy to approximate the effect of averaging the predictions of all these thinned networks by simply using a single unthinned network that has smaller weights. This significantly reduces overfitting and gives major improvements over other regularization methods. We show that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.

/pdf/dropout-a-simple-way-to-prevent-neural-networks-from-4gtr92okhv.pdf

Dropout: a simple way to prevent neural networks from overfitting

抗原变异可使得多种致病微生物易于逃避宿主免疫应答。表达在感染红细胞表面的恶性疟原虫红细胞表面蛋白1（PfPMP1）与感染红细胞、内皮细胞、树突状细胞以及胎盘的单个或多个受体作用，在黏附及免疫逃避中起关键的作用。每个单倍体基因组var基因家族编码约60种成员，通过启动转录不同的var基因变异体为抗原变异提供了分子基础。

疟原虫var基因转换速率变化导致抗原变异[英]／Paul H, Robert P, Christodoulou Z, et al//Proc Natl Acad Sci U S A

The identification of genetically homogeneous groups of individuals is a long standing issue in population genetics. A recent Bayesian algorithm implemented in the software STRUCTURE allows the identification of such groups. However, the ability of this algorithm to detect the true number of clusters (K) in a sample of individuals when patterns of dispersal among populations are not homogeneous has not been tested. The goal of this study is to carry out such tests, using various dispersal scenarios from data generated with an individual-based model. We found that in most cases the estimated 'log probability of data' does not provide a correct estimation of the number of clusters, K. However, using an ad hoc statistic DeltaK based on the rate of change in the log probability of data between successive K values, we found that STRUCTURE accurately detects the uppermost hierarchical level of structure for the scenarios we tested. As might be expected, the results are sensitive to the type of genetic marker used (AFLP vs. microsatellite), the number of loci scored, the number of populations sampled, and the number of individuals typed in each sample.

/pdf/detecting-the-number-of-clusters-of-individuals-using-the-16mlx2qlv8.pdf

Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study.

Arlequin ver 3.0 is a software package integrating several basic and advanced methods for population genetics data analysis, like the computation of standard genetic diversity indices, the estimation of allele and haplotype frequencies, tests of departure from linkage equilibrium, departure from selective neutrality and demographic equilibrium, estimation or parameters from past population expansions, and thorough analyses of population subdivision under the AMOVA framework. Arlequin 3 introduces a completely new graphical interface written in C++, a more robust semantic analysis of input files, and two new methods: a Bayesian estimation of gametic phase from multi-locus genotypes, and an estimation of the parameters of an instantaneous spatial expansion from DNA sequence polymorphism. Arlequin can handle several data types like DNA sequences, microsatellite data, or standard multi-locus genotypes. A Windows version of the software is freely available on http://cmpg.unibe.ch/software/arlequin3.

/pdf/arlequin-version-3-0-an-integrated-software-package-for-4cs6nd3y2p.pdf

Arlequin (version 3.0): An integrated software package for population genetics data analysis

We present here a new version of the Arlequin program available under three different forms: a Windows graphical version (Winarl35), a console version of Arlequin (arlecore), and a specific console version to compute summary statistics (arlsumstat). The command-line versions run under both Linux and Windows. The main innovations of the new version include enhanced outputs in XML format, the possibility to embed graphics displaying computation results directly into output files, and the implementation of a new method to detect loci under selection from genome scans. Command-line versions are designed to handle large series of files, and arlsumstat can be used to generate summary statistics from simulated data sets within an Approximate Bayesian Computation framework.

Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows

Models for the genetic evolution of natural populations have supplied the inspiration for the adaptive computer algorithms known as “genetic algorithms.” In its original form, a genetic algorithm simulates the evolution of a haploid population: Genetic variation is produced by mutation at a number of genetic loci in a chromosome, represented as a string of bits, and the variants in the various chromosomes are reshuffled by genetic recombination. In nature, the ability of a haploid individual to survive and reproduce is expressed as its fitness; in computer science the value of a string is measured by its ability to program a given task. The performance of genetic algorithms is evaluated in the “schemata theorem,” which we present and discuss in the context of the population genetics of multiple loci and propose a generalization of the theorem. © 1998 John Wiley & Sons, Inc.

Algorithms, genetics, and populations: the schemata theorem revisited

Within the blink of an eye on a geological timescale, humans advanced from using basic stone tools to examining the rocks on Mars; however, our exact evolutionary path and the relative importance of genetic and cultural evolution in directing it remain a mystery. Our cultural capacities—to generate new ideas, to communicate and learn from one another, and to form vast social networks—together make us uniquely human, but the origins, the mechanisms and the evolutionary impact of these capacities are not well understood.

This special issue comprises studies that bring together perspectives from anthropology, archaeology, biology, computer science, ecology and psychology to help elucidate the cultural forces affecting human evolution. These studies explore avenues in which approaches and insights from different fields may inform one another or be brought together to generate novel interdisciplinary research agendas, with the goal of advancing the study of human uniqueness and its pre-hominid origins. The aim of this issue is to advance interdisciplinary discussion of the roles that culture plays in shaping the course of human evolution, exploring the mechanisms of cultural evolution from their cognitive underpinnings in individuals, through the behavioural ecology of learning from others, to the dynamics of transmission at the level of individuals and populations. The articles in this issue bring insights from disparate disciplines to bear on major questions in cultural evolution [1–11] and suggest broad-scale ways in which the study of cultural evolution can be synthesized with other disciplines [12–14].

In the introduction to this issue [15], we outline how integrative studies are poised to move the field of cultural evolution forward; we demonstrate the utility of this …

https://royalsocietypublishing.org/doi/pdf/10.1098/rstb.2017.0413

Bridging cultural gaps: interdisciplinary studies in human cultural evolution.

Static environments with limited resources select for multiple foraging strategies rather than conformity

Ii. differentiation among neighboring populations

Optimizing average path length (APL) by adding shortcut edges has been widely discussed in connection with social networks, but the relationship between network diameter and APL is generally ignored in the dynamic optimization of APL. In this paper, we analyze this relationship and transform the problem of optimizing APL into the problem of decreasing diameter to 2. We propose a mathematic model based on a memetic algorithm. Experimental results show that our algorithm can efficiently solve this problem as well as optimize APL.

/pdf/optimization-of-the-critical-diameter-and-average-path-53gtpy02xu.pdf

Marcus W. Feldman

Papers

Algorithms, genetics, and populations: the schemata theorem revisited

Bridging cultural gaps: interdisciplinary studies in human cultural evolution.

Static environments with limited resources select for multiple foraging strategies rather than conformity

Ii. differentiation among neighboring populations

Optimization of the Critical Diameter and Average Path Length of Social Networks