National Repository of Grey Literature 2 records found  Search took 0.01 seconds. 
Grafové neuronové sítě pro odhad výkonnosti při hledání architektur
Suchopárová, Gabriela ; Neruda, Roman (advisor) ; Pilát, Martin (referee)
In this work we present a novel approach to network embedding for neural architecture search - info-NAS. The model learns to predict the output fea- tures of a trained convolutional neural network on a set of input images. We use the NAS-Bench-101 search space as the neural architecture dataset, and the CIFAR-10 as the image dataset. For the purpose of this task, we extend an existing unsupervised graph variational autoencoder, arch2vec, by jointly training on unlabeled and labeled neural architectures in a semi-supervised manner. To evaluate our approach, we analyze how our model learns on the data, compare it to the original arch2vec, and finally, we evaluate both mod- els on the NAS-Bench-101 search task and on the performance prediction task. 1
Evolutionary optimization of machine learning workflows
Suchopárová, Gabriela ; Neruda, Roman (advisor) ; Pilát, Martin (referee)
This work deals with automated machine learning (AutoML), which is a field that aims to automatize the process of model selection for a given machine learning problem. We have developed a system that, for a given supervised learning task represented by a dataset, finds a suitable pipeline - combination of machine learning, ensembles and preprocessing methods. For the search we designed a special instance of the developmental genetic programming which enables us to encode directed acyclic graph pipelines into a tree representation. The system is implemented in the Python programming language and operates on top of the scikit-learn library. The performance of our solution was tested on 72 datasets of the OpenML-CC18 benchmark with very good results. 1

Interested in being notified about new results for this query?
Subscribe to the RSS feed.