Neural networks are susceptible to minor distortions in their input, which can lead to errors they would not otherwise make. This susceptibility, termed as the network’s robustness, is a crucial...Show moreNeural networks are susceptible to minor distortions in their input, which can lead to errors they would not otherwise make. This susceptibility, termed as the network’s robustness, is a crucial aspect to evaluate. While several methods exist for measuring robustness, they usually suffer from interpretability issues and do not provide a statistical guarantee. In this work, we propose a novel robustness measure that addresses these short- comings by modeling the robustness as a probability distribution and mea- suring its 0.05 quantile. Additionally, previous work suggests the poten- tial modeling of robustness through a log-normal distribution. To eval- uate this hypothesis and its computational benefits, we introduce an es- timator that assumes the distribution is log-normal. A comparison with the standard parameter-free estimator reveals significantly improved com- putational efficiency with the parametrized approach. However, the log- normal assumption requires further research. The assumption is too strong and needs to be relaxed before the parametrized estimator can reliably be utilized.Show less
Discrimination and prejudice are serious problems in our society. The desire for cognitive consistency is eminently influential on how such implicit and explicit attitudes translate into behavior...Show moreDiscrimination and prejudice are serious problems in our society. The desire for cognitive consistency is eminently influential on how such implicit and explicit attitudes translate into behavior and how attitudes change over different contexts. While it has been found that experiencing cognitive dissonance between discriminating behaviors and one’s values can lead to bias reduction, a recent study by Szekeres et al. (2022) also found the opposite: Inaction, or a failure to confront, in the face of witnessing prejudice, subsequently correlates with more discriminating attitudes. This effect was reported to be mediated by how much a person values confronting; it was specifically observed for respondents that valued confronting highly. The present paper aimed to test the robustness of this interaction of confronting importance with confronting behavior on attitude change across other subsamples. For this, three different resampling approaches instrumentalizing bootstrapping were employed that excluded different percentages of random respondents. The exclusions were either completely random or random within either extreme or central scoring participants. The comparison of the mean interaction coefficients and variances of the different resampling conditions showed that the mean interaction remained quite stable after resampling. Only excluding large percentages of extreme values led to somewhat of a drift to a larger interaction effect.Show less
Statistical hypothesis testing is central to many scientific fields. Testing many hypothe-ses simultaneously is called multiple testing. The main concern in multiple testing, is toensure that most...Show moreStatistical hypothesis testing is central to many scientific fields. Testing many hypothe-ses simultaneously is called multiple testing. The main concern in multiple testing, is toensure that most of the rejected null hypotheses are indeed false, i.e., that the numberof incorrect rejections remains low. A major challenge in multiple testing is to accountfor the complex dependencies in the data. A powerful approach in this regard, arepermutation-based multiple testing methods. These methods make few distributionalassumptions. In fact, they often make only one assumption, called joint exchangeabil-ity. In this thesis we investigate the robustness of the methods to violations of thisassumption. We do this by means of simulations, where we focus on case-control data.We find that, while the theoretical literature always makes the mentioned assumption,it is often not necessary in practice. Thus, this thesis provides further evidence for thevalidity of these powerful methods in practice.Show less
Statistical hypothesis testing is central to many scientific fields. Testing many hypotheses simultaneously is called multiple testing. The main concern in multiple testing, is to ensure that most...Show moreStatistical hypothesis testing is central to many scientific fields. Testing many hypotheses simultaneously is called multiple testing. The main concern in multiple testing, is to ensure that most of the rejected null hypotheses are indeed false, i.e., that the number of incorrect rejections remains low. A major challenge in multiple testing is to account for the complex dependencies in the data. A powerful approach in this regard, are permutation-based multiple testing methods. These methods make few distributional assumptions. In fact, they often make only one assumption, called joint exchangeability. In this thesis we investigate the robustness of the methods to violations of this assumption. We do this by means of simulations, where we focus on case-control data. We find that, while the theoretical literature always makes the mentioned assumption, it is often not necessary in practice. Thus, this thesis provides further evidence for the validity of these powerful methods in practice.Show less
This research set out to determine whether or not a distinction could be made between three different kinds of (pre)historic societies, namely terrestrial hunter-gatherers, marine hunter-gatherers...Show moreThis research set out to determine whether or not a distinction could be made between three different kinds of (pre)historic societies, namely terrestrial hunter-gatherers, marine hunter-gatherers and agriculturalists based on the level of activity. What was examined was that, within terrestrial hunter-gatherers, the lower limb showed greater diaphyseal strength due to long-distance traveling and running. Within the marine hunter-gatherers, the upper limb showed greater diaphyseal strength due to the fact that their way of transportation would mainly involve either swimming or the use of watercrafts. Also they do not need to be as nomadic as terrestrial hunter-gatherers. Both terrestrial and marine hunter-gatherers show a great level of sexual dimorphism within their skeleton due to different jobs performed by men and women. Within agricultural societies there is a decrease in diaphyseal strength of the long bones, whether these are from the upper limb or from the lower limb, and their shape changes from more oval to more circular. Another notorious change witnessed within the transition to agriculture is that there is a decrease in sexual dimorphism. This topic is still in a primal phase of investigation and telling activity patterns from bone is still very difficult.Show less
