Neural Architecture Search is an active research field that aims to design neural networks automatically. Nevertheless, this is usually an expensive process since the search algorithm must evaluate the performance of many candidate solutions from a vast search space. Because of that, different strategies have been proposed to perform efficient Neural Architecture Search. The recent development of zero-cost performance predictors has shown a lot of promise due to the possibility of predicting a network's performance without training. On the other hand, a predictor's correlation with a model's performance may depend on the network search space and on the benchmark dataset. Each performance predictor might lead the search processes to favor very different network patterns. A design principle is defined as a restriction in a hyperparameter distribution that is expected to yield optimum network performance. In this work, we propose an automatic iterative approach to uncover the design principles of deep neural networks optimized by zero-cost performance predictors, and we discuss insightful information obtained by its application.

Neural Architecture Search (NAS) is a busy research field growing exponentially in recent years. State-of-the-art deep neural networks usually require a specialist to fine-tune the model to solve a specific problem. NAS research aims to design neural network architectures automatically, thus easing the need for machine learning specialists to spend a lot of effort on hand-crafted attempts. As artificial intelligence applications are becoming ubiquitous, there is also a growing interest in efficient applications that could be deployed to smartphones, smart wearable devices, and other edge devices. Gender recognition in unfiltered images — such as those we find in real-world situations like pictures taken with smartphones and video shots from surveillance cameras — is one of such challenging applications. In this work, we developed an evolutionary NAS algorithm that consistently finds efficient ResNet-based architectures, named RENNAS, which have a good trade-off between classification accuracy and architectural and computational complexities. We demonstrate our algorithm's performance on Adience dataset of unfiltered images for gender recognition.

MOTIVATION: We present the first tool for unbiased quality control of top-down proteomics datasets. Our tool can select high-quality top-down proteomics spectra, serve as a gateway for building top-down spectral libraries and, ultimately, improve identification rates. RESULTS: We demonstrate that a twofold rate increase for two E. coli top-down proteomics datasets may be achievable. AVAILABILITY AND IMPLEMENTATION: http://patternlabforproteomics.org/tdgc, freely available for academic use. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Around 75% of all mass spectra remain unidentified by widely adopted proteomic strategies. We present DiagnoProt, an integrated computational environment that can efficiently cluster millions of spectra and use machine learning to shortlist high-quality unidentified mass spectra that are discriminative of different biological conditions. RESULTS: We exemplify the use of DiagnoProt by shortlisting 4366 high-quality unidentified tandem mass spectra that are discriminative of different types of the Aspergillus fungus. AVAILABILITY AND IMPLEMENTATION: DiagnoProt, a demonstration video and a user tutorial are available at http://patternlabforproteomics.org/diagnoprot . CONTACT: [email protected] or [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.