The field of Automated Machine Learning (AutoML) has as its main goal to automate the process of creating complete Machine Learning (ML) pipelines to any dataset without requiring deep user expertise in ML. Several AutoML methods have been proposed so far, but there is not a single one that really stands out. Furthermore, there is a lack of studies on the characteristics of the fitness landscape of AutoML search spaces. Such analysis may help to understand the performance of different optimization methods for AutoML and how to improve them. This paper adapts classic fitness landscape analysis measures to the context of AutoML. This is a challenging task, as AutoML search spaces include discrete, continuous, categorical and conditional hyperparameters. We propose an ML pipeline representation, a neighborhood definition and a distance metric between pipelines, and use them in the evaluation of the fitness distance correlation (FDC) and the neutrality ratio for a given AutoML search space. Results of FDC are counter-intuitive and require a more in-depth analysis of a range of search spaces. Results of neutrality, in turn, show a strong positive correlation between the mean neutrality ratio and the fitness value.

No abstract

This supplementary material aims to describe the proposed multilabel classification (MLC) search spaces based on the MEKA and WEKA softwares. First, we overview 26 MLC algorithms and metaalgorithms in MEKA, presenting their main characteristics, such as hyper-parameters, dependencies and constraints. Second, we review 28 single-label classification (SLC) algorithms, preprocessing algorithms and meta-algorithms in the WEKA software. These SLC algorithms were also studied because they are part of the proposed MLC search spaces. Fundamentally, this occurs due to the problem transformation nature of several MLC algorithms used in this work. These algorithms transform an MLC problem into one or several SLC problems in the first place and solve them with SLC model(s) in a next step. Therefore, understanding their main characteristics is crucial to this work. Finally, we present a formal description of the search spaces by proposing a context-free grammar that encompasses the 54 learning algorithms. This grammar basically comprehends the possible combinations, the constraints and dependencies among the learning algorithms.

The widely used herbicide atrazine is a potent endocrine disruptor known to cause increased aromatase expression and transient increase in testicular weight followed by remarkable testis atrophy. However, whether the effects of atrazine on the testes are primary or secondary to dysfunctions in other components of male reproductive tract remains unknown. Given the high sensitivity of the efferent ductules to estrogen imbalance and the similarity to alterations previously described for other disruptors of these ductules function, and the testicular alterations observed after atrazine exposure, we hypothesized that the efferent ductules could be a target for atrazine. Herein we characterized the efferent ductules and the ventral prostate of adult Wistar rats treated with 200 mg/kg/day of atrazine for 7, 15, and 40 days. Additionally, we evaluated if the effects of atrazine in these organs could be reduced after discontinuation of the treatment. Atrazine exposure resulted in mild effects on the ventral prostate, but remarkable alterations on the efferent ductules, including luminal dilation, reduced epithelial height, and disruption of the epithelial homeostasis, which coincides with increased aromatase expression. Together with our previous data, these results suggest that at least part of the testicular effects of atrazine may be secondary to the alterations in the efferent ductules.

No abstract