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
Correlation Congruence for Knowledge Distillation
TL;DR: A new framework named correlation congruence for knowledge distillation (CCKD), which transfers not only the instance-level information but also the correlation between instances, and a generalized kernel method based on Taylor series expansion is proposed to better capture the correlationBetween instances.
Proceedings ArticleDOI
Learning Shape Templates With Structured Implicit Functions
TL;DR: It is shown that structured implicit functions are suitable for learning and allow a network to smoothly and simultaneously fit multiple classes of shapes in a general shape template from data.
Proceedings Article
Adaptive neural networks for efficient inference
TL;DR: It is shown that computational time can be dramatically reduced by exploiting the fact that many examples can be correctly classified using relatively efficient networks and that complex, computationally costly networks are only necessary for a small fraction of examples.
Proceedings ArticleDOI
Understanding Deep Networks via Extremal Perturbations and Smooth Masks
TL;DR: Some of the shortcomings of existing approaches to perturbation analysis are discussed and the concept of extremal perturbations are introduced, which are theoretically grounded and interpretable and allow us to remove all tunable weighing factors from the optimization problem.
Proceedings ArticleDOI
Learning with Side Information through Modality Hallucination
TL;DR: This work presents a modality hallucination architecture for training an RGB object detection model which incorporates depth side information at training time and transfers information commonly extracted from depth training data to a network which can extract that information from the RGB counterpart.
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.