L
Laurens van der Maaten
Researcher at Facebook
Publications - 127
Citations - 79845
Laurens van der Maaten is an academic researcher from Facebook. The author has contributed to research in topics: Computer science & Network architecture. The author has an hindex of 47, co-authored 118 publications receiving 54188 citations. Previous affiliations of Laurens van der Maaten include Maastricht University & Delft University of Technology.
Papers
More filters
Journal ArticleDOI
Convolutional Networks with Dense Connectivity
TL;DR: DenseNet as discussed by the authors proposes to connect each layer to every other layer in a feed-forward fashion, which alleviates the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially improve parameter efficiency.
Posted Content
Submanifold Sparse Convolutional Networks
TL;DR: This work introduces a sparse convolutional operation tailored to processing sparse data that operates strictly on submanifolds, rather than "dilating" the observation with every layer in the network.
Proceedings Article
Countering Adversarial Images using Input Transformations
TL;DR: In this paper, the authors apply image transformations such as bit-depth reduction, JPEG compression, total variance minimization, and image quilting before feeding the image to a convolutional network classifier.
Book ChapterDOI
Revisiting Visual Question Answering Baselines
TL;DR: The authors proposed a simple alternative model based on binary classification, which receives the answer as input and predicts whether or not an image-question-answer triplet is correct, which achieves state-of-the-art performance on the Visual7W Telling task and compares surprisingly well with the most complex systems proposed for the VQA Real Multiple Choice task.