Open AccessProceedings Article
Selective Classification for Deep Neural Networks
Yonatan Geifman,Ran El-Yaniv +1 more
- Vol. 30, pp 4878-4887
Reads0
Chats0
TLDR
A method to construct a selective classifier given a trained neural network, which allows a user to set a desired risk level and the classifier rejects instances as needed, to grant the desired risk (with high probability).Abstract:
Selective classification techniques (also known as reject option) have not yet been considered in the context of deep neural networks (DNNs). These techniques can potentially significantly improve DNNs prediction performance by trading-off coverage. In this paper we propose a method to construct a selective classifier given a trained neural network. Our method allows a user to set a desired risk level. At test time, the classifier rejects instances as needed, to grant the desired risk (with high probability). Empirical results over CIFAR and ImageNet convincingly demonstrate the viability of our method, which opens up possibilities to operate DNNs in mission-critical applications. For example, using our method an unprecedented 2% error in top-5 ImageNet classification can be guaranteed with probability 99.9%, with almost 60% test coverage.read more
Citations
More filters
Proceedings Article
Generative Well-intentioned Networks
Justin Cosentino,Jun Zhu +1 more
TL;DR: Generative Well-intentioned Networks are proposed, a novel framework for increasing the accuracy of certainty-based, closed-world classifiers by introducing a reject option to the classifier during inference, allowing the classifiers to reject an observation instance rather than predict an uncertain label.
Proceedings Article
Budget Learning via Bracketing
TL;DR: This work proposes a new formulation for the budget learning problem via the concept of bracketings, providing PAC-style learnability definitions; associating the notion of budget learnability to approximability via brackets; and giving VC-theoretic analyses of their properties.
Posted Content
Utilizing Network Properties to Detect Erroneous Inputs.
Matt Gorbett,Nathaniel Blanchard +1 more
TL;DR: This work trains a linear SVM classifier to detect these four types of erroneous data using hidden and softmax feature vectors of pre-trained neural networks, and indicates that these faulty data types generally exhibit linearly separable activation properties from correct examples, giving the ability to reject bad inputs with no extra training or overhead.
Journal Article
SoCal: Selective Oracle Questioning for Consistency-based Active Learning of Cardiac Signals
TL;DR: This work proposes SoCal, a consistency-based AL framework that dynamically determines whether to request a label from an oracle or to generate a pseudo-label instead, and shows that this framework decreases the labelling burden while maintaining strong performance, even in the presence of a noisy oracle.
Journal ArticleDOI
Stop Overcomplicating Selective Classification: Use Max-Logit
TL;DR: This paper argues that the superior performance of state-of-the-art methods is owed to training a more generalizable classifier; however, their selection mechanism is suboptimal, and motivates an alternative selection strategy based on the cross entropy loss for the classi-class settings, namely, max of the logits.
References
More filters
Proceedings ArticleDOI
Deep Residual Learning for Image Recognition
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Journal ArticleDOI
ImageNet Large Scale Visual Recognition Challenge
Olga Russakovsky,Jia Deng,Hao Su,Jonathan Krause,Sanjeev Satheesh,Sean Ma,Zhiheng Huang,Andrej Karpathy,Aditya Khosla,Michael S. Bernstein,Alexander C. Berg,Li Fei-Fei +11 more
TL;DR: The ImageNet Large Scale Visual Recognition Challenge (ILSVRC) as mentioned in this paper is a benchmark in object category classification and detection on hundreds of object categories and millions of images, which has been run annually from 2010 to present, attracting participation from more than fifty institutions.
Proceedings ArticleDOI
Very deep convolutional neural network based image classification using small training sample size
Shuying Liu,Weihong Deng +1 more
TL;DR: In this article, a modified VGG-16 network was used to fit CIFAR-10 without severe overfitting and achieved 8.45% error rate on the dataset.
Journal ArticleDOI
An optimum character recognition system using decision functions
TL;DR: The character recognition problem, usually resulting from characters being corrupted by printing deterioration and/or inherent noise of the devices, is considered from the viewpoint of statistical decision theory and the optimum recogition is obtained.