590 citations · 2.3k across the 19 of their papers we have counts for
7 papers · 1 filter
When adversarial examples are excusable
Pieter-Jan Kindermans, Charles Staats
Neural networks work remarkably well in practice and theoretically they can be universal approximators. However, they still make mistakes and a specific type of them called adversa…
TabNAS: Rejection Sampling for Neural Architecture Search on Tabular Datasets
Chengrun Yang, Gabriel Bender, Hanxiao Liu +5
The best neural architecture for a given machine learning problem depends on many factors: not only the complexity and structure of the dataset, but also on resource constraints in…
Can weight sharing outperform random architecture search? An investigation with TuNAS
Gabriel Bender, Hanxiao Liu, Bo Chen +4
Efficient Neural Architecture Search methods based on weight sharing have shown good promise in democratizing Neural Architecture Search for computer vision models. There is, howev…
Neural Predictor for Neural Architecture Search
Wei Wen, Hanxiao Liu, Hai Li +3
Neural Architecture Search methods are effective but often use complex algorithms to come up with the best architecture. We propose an approach with three basic steps that is conce…
iNNvestigate neural networks!
Maximilian Alber, Sebastian Lapuschkin, Philipp Seegerer +7
In recent years, deep neural networks have revolutionized many application domains of machine learning and are key components of many critical decision or predictive processes. The…
A Benchmark for Interpretability Methods in Deep Neural Networks
Sara Hooker, Dumitru Erhan, Pieter-Jan Kindermans +1
We propose an empirical measure of the approximate accuracy of feature importance estimates in deep neural networks. Our results across several large-scale image classification dat…
Don't Decay the Learning Rate, Increase the Batch Size
Samuel L. Smith, Pieter-Jan Kindermans, Chris Ying +1
It is common practice to decay the learning rate. Here we show one can usually obtain the same learning curve on both training and test sets by instead increasing the batch size du…