Towards Open Set Deep Networks

Abstract

Deep networks have produced significant gains for various visual recognition problems, leading to high impact academic and commercial applications. Recent work in deep networks highlighted that it is easy to generate images that humans would never classify as a particular object class, yet networks classify such images high confidence as that given class -deep network are easily fooled with images humans do not consider meaningful. The closed set nature of deep networks forces them to choose from one of the known classes leading to such artifacts. Recognition in the real world is open set, i.e. the recognition system should reject unknown/unseen classes at test time. We present a methodology to adapt deep networks for open set recognition, by introducing a new model layer, OpenMax, which estimates the probability of an input being from an unknown class. A key element of estimating the unknown probability is adapting Meta-Recognition concepts to the activation patterns in the penultimate layer of the network. Open-Max allows rejection of "fooling" and unrelated open set images presented to the system; OpenMax greatly reduces the number of obvious errors made by a deep network. We prove that the OpenMax concept provides bounded open space risk, thereby formally providing an open set recognition solution. We evaluate the resulting open set deep networks using pre-trained networks from the Caffe Model-zoo on ImageNet 2012 validation data, and thousands of fooling and open set images. The proposed OpenMax model significantly outperforms open set recognition accuracy of basic deep networks as well as deep networks with thresholding of SoftMax probabilities.

BranchyNet: Fast inference via early exiting from deep neural networks

“…The MLP at an exit point would attempt to predict if a given sample would be correctly classified at the specific exit. More generally, this approach is closely related to the open world recognition problem [2], [1], which is interested in quantifying the uncertainty of a model for a particular set of unseen or out of set test samples. We can expand on the MLP approach further by using a different formulation than SoftMax, such as OpenMax [2], which attempts to quantify the uncertainty directly in the probability vectorŷ by adding an additional uncertain class.…”

CHEER: hierarCHical taxonomic classification for viral mEtagEnomic data via deep leaRning

“…However, for an irrelevant input from an unknown clade, all classes in the model tend to have low probabilities and thus applying a threshold on uncertainty can be used to reject unknown classes [25]. Along with this idea, serval approaches [28,29] have been proposed to solve the open set problem.…”

DOC: Deep Open Classification of Text Documents

“…OpenMax: This is the latest method from computer vision (Bendale and Boult, 2016). Since it is a CNN-based method for image classification, we adapt it for text classification by using CNN with a softmax output layer, and adopt the OpenMax layer 3 for open text classification.…”

“…Traditional multi-class classifiers (Goodfellow et al, 2016;Bendale and Boult, 2016) typically use softmax as the final output layer, which does not have the rejection capability since the probability of prediction for each class is normalized across all training/seen classes. Instead, we build a 1-vs-rest layer containing m sigmoid functions for m seen classes.…”

“…The work closest to ours is that in (Bendale and Boult, 2016), which leverages an algorithm called OpenMax to add the rejection capability by utilizing the logits that are trained via closed-world softmax function. One weak assumption of OpenMax is that examples with equally likely logits are more likely from the unseen or rejection class, which can be examples that are hard to classify.…”