Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
Reads0
Chats0
TLDR
Virtual adversarial training (VAT) as discussed by the authors is a regularization method based on virtual adversarial loss, which is a measure of local smoothness of the conditional label distribution given input.Abstract:
We propose a new regularization method based on virtual adversarial loss: a new measure of local smoothness of the conditional label distribution given input. Virtual adversarial loss is defined as the robustness of the conditional label distribution around each input data point against local perturbation. Unlike adversarial training, our method defines the adversarial direction without label information and is hence applicable to semi-supervised learning. Because the directions in which we smooth the model are only “virtually” adversarial, we call our method virtual adversarial training (VAT). The computational cost of VAT is relatively low. For neural networks, the approximated gradient of virtual adversarial loss can be computed with no more than two pairs of forward- and back-propagations. In our experiments, we applied VAT to supervised and semi-supervised learning tasks on multiple benchmark datasets. With a simple enhancement of the algorithm based on the entropy minimization principle, our VAT achieves state-of-the-art performance for semi-supervised learning tasks on SVHN and CIFAR-10.read more
Citations
More filters
Proceedings ArticleDOI
Virtual Adversarial Training for DS-CNN Based Small-Footprint Keyword Spotting
Xiong Wang,Sining Sun,Lei Xie +2 more
TL;DR: This paper proposes to train a DS-CNN KWS model using adversarial regularization, which aims to smooth model's distribution and thus to improve robustness, by explicitly introducing a distribution smoothness measure into the loss function.
Posted ContentDOI
Meta-Learning Biologically Plausible Semi-Supervised Update Rules
TL;DR: Meta-learning is believed to be a powerful approach to finding more effective synaptic plasticity rules, which will motivate new hypotheses for biological neural networks, and new algorithms for artificial neural networks.
Journal ArticleDOI
Semi-Supervised Image Deraining Using Gaussian Processes
TL;DR: In this paper, a Gaussian Process-based semi-supervised learning framework is proposed to enable the network in learning to derain using synthetic dataset while generalizing better using unlabeled real-world images.
Proceedings ArticleDOI
RECORD: Resource Constrained Semi-Supervised Learning under Distribution Shift
Lan-Zhe Guo,Zhi Zhou,Yu-Feng Li +2 more
TL;DR: A novel influence-based approach is used to select the most influential samples for the distribution change based on resource constraints and free up memory to put the latest unlabeled data with its pseudo-label for the next distribution tracking.
Does Adversarial Transferability Indicate Knowledge Transferability
TL;DR: It is shown that composition with an affine function is sufficient to reduce the difference between two models when adversarialtransferability between them is high, and a strong positive correlation between the adversarial transferability and knowledge transferability is observed.
References
More filters
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Journal ArticleDOI
Generative Adversarial Nets
Ian Goodfellow,Jean Pouget-Abadie,Mehdi Mirza,Bing Xu,David Warde-Farley,Sherjil Ozair,Aaron Courville,Yoshua Bengio +7 more
TL;DR: A new framework for estimating generative models via an adversarial process, in which two models are simultaneously train: a generative model G that captures the data distribution and a discriminative model D that estimates the probability that a sample came from the training data rather than G.
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.
Proceedings Article
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Sergey Ioffe,Christian Szegedy +1 more
TL;DR: Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.
Proceedings ArticleDOI
Densely Connected Convolutional Networks
TL;DR: DenseNet as mentioned in this paper proposes to connect each layer to every other layer in a feed-forward fashion, which can alleviate the vanishing gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters.