Open AccessPosted Content
Distilling the Knowledge in a Neural Network
Reads0
Chats0
TLDR
This work shows that it can significantly improve the acoustic model of a heavily used commercial system by distilling the knowledge in an ensemble of models into a single model and introduces a new type of ensemble composed of one or more full models and many specialist models which learn to distinguish fine-grained classes that the full models confuse.Abstract:
A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions. Unfortunately, making predictions using a whole ensemble of models is cumbersome and may be too computationally expensive to allow deployment to a large number of users, especially if the individual models are large neural nets. Caruana and his collaborators have shown that it is possible to compress the knowledge in an ensemble into a single model which is much easier to deploy and we develop this approach further using a different compression technique. We achieve some surprising results on MNIST and we show that we can significantly improve the acoustic model of a heavily used commercial system by distilling the knowledge in an ensemble of models into a single model. We also introduce a new type of ensemble composed of one or more full models and many specialist models which learn to distinguish fine-grained classes that the full models confuse. Unlike a mixture of experts, these specialist models can be trained rapidly and in parallel.read more
Citations
More filters
Proceedings ArticleDOI
Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells
TL;DR: This work focuses on dense per-pixel tasks, in particular, semantic image segmentation using fully convolutional networks, and relies on a progressive strategy that terminates non-promising architectures from being further trained, and on Polyak averaging coupled with knowledge distillation to speed-up the convergence.
Posted Content
Long-tailed Recognition by Routing Diverse Distribution-Aware Experts
TL;DR: RIDE aims to reduce both the bias and the variance of a long-tailed classifier by RoutIng Diverse Experts (RIDE), which significantly outperforms the state-of-the-art methods by 5% to 7% on all the benchmarks including CIFAR100-LT, ImageNet-LT and iNaturalist.
Posted ContentDOI
Using Deep Learning to Annotate the Protein Universe
Maxwell L. Bileschi,David Belanger,Drew Bryant,Theo Sanderson,Brandon Carter,D. Sculley,Mark A. DePristo,Lucy J. Colwell,Lucy J. Colwell +8 more
TL;DR: This paper explores an alternative methodology based on deep learning that learns the relationship between unaligned amino acid sequences and their functional annotations across all 17929 families of the Pfam database, and reports convolutional networks that are significantly more accurate and computationally efficient than BLASTp.
Posted Content
Are All Layers Created Equal
TL;DR: This study provides further evidence that mere parameter counting or norm accounting is too coarse in studying generalization of deep models, and flatness or robustness analysis of the models needs to respect the network architectures.
Proceedings ArticleDOI
Mobile Video Object Detection with Temporally-Aware Feature Maps
Menglong Zhu,Mason Liu +1 more
TL;DR: This approach combines fast single-image object detection with convolutional long short term memory layers to create an inter-weaved recurrent-convolutional architecture that is substantially faster than existing detection methods in video and significantly reduces computational cost.
References
More filters
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Journal ArticleDOI
Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups
Geoffrey E. Hinton,Li Deng,Dong Yu,George E. Dahl,Abdelrahman Mohamed,Navdeep Jaitly,Andrew W. Senior,Vincent Vanhoucke,Patrick Nguyen,Tara N. Sainath,Brian Kingsbury +10 more
TL;DR: This article provides an overview of progress and represents the shared views of four research groups that have had recent successes in using DNNs for acoustic modeling in speech recognition.
Posted Content
Improving neural networks by preventing co-adaptation of feature detectors
TL;DR: The authors randomly omits half of the feature detectors on each training case to prevent complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors.
Book ChapterDOI
Ensemble Methods in Machine Learning
TL;DR: Some previous studies comparing ensemble methods are reviewed, and some new experiments are presented to uncover the reasons that Adaboost does not overfit rapidly.