Multiple Class Novelty Detection Under Data Distribution Shift

Oza, Poojan; Nguyen, Hien Van; Patel, Vishal M.

doi:10.1007/978-3-030-58571-6_26

Cited by 16 publications

(8 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the former, explicit relations between regions like visual relation detection [30,31,33] may improve the diagram understanding. For the latter, fine grained attentions may enhance the reasoning ability to overcome the data shift [34,35]. Furthermore, we conduct the pair-wise significance test (paired t-test) between WSTQ * and ISAAQ * on each subtask.…”

Section: B Results On Ck12-qamentioning

confidence: 99%

Weakly Supervised Learning for Textbook Question Answering

Ma¹,

Chai²,

Huang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Textbook Question Answering (TQA) is the task of answering diagram and non-diagram questions given large multi-modal contexts consisting of abundant text and diagrams. Deep text understandings and effective learning of diagram semantics are important for this task due to its specificity. In this paper, we propose a Weakly Supervised learning method for TQA (WSTQ), which regards the incompletely accurate results of essential intermediate procedures for this task as supervision to develop Text Matching (TM) and Relation Detection (RD) tasks and then employs the tasks to motivate itself to learn strong text comprehension and excellent diagram semantics respectively. Specifically, we apply the result of text retrieval to build positive as well as negative text pairs. In order to learn deep text understandings, we first pre-train the text understanding module of WSTQ on TM and then fine-tune it on TQA. We build positive as well as negative relation pairs by checking whether there is any overlap between the items/regions detected from diagrams using object detection. The RD task forces our method to learn the relationships between regions, which are crucial to express the diagram semantics. We train WSTQ on RD and TQA simultaneously, \emph{i.e.}, multitask learning, to obtain effective diagram semantics and then improve the TQA performance. Extensive experiments are carried out on CK12-QA and AI2D to verify the effectiveness of WSTQ. Experimental results show that our method achieves significant accuracy improvements of $5.02\%$ and $4.12\%$ on test splits of the above datasets respectively than the current state-of-the-art baseline. We have released our code on \url{https://github.com/dr-majie/WSTQ}.

show abstract

Section: B Results On Ck12-qamentioning

confidence: 99%

Weakly Supervised Learning for Textbook Question Answering

Ma¹,

Chai²,

Huang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Additionally, human relabeling is an expensive and time-consuming solution. There are more works using novelty detector for outlier detections (Abati et al [50], Oza et al [51], Lee et al [52], Perera et al [53], Sabokrou et al [54], Chen et al [55], Zhu et al [56]). Erichson et al [57] studies the behavior of the backdoored network using noise response analysis.…”

Section: Related Workmentioning

confidence: 99%

A Feature-Based On-Line Detector to Remove Adversarial-Backdoors by Iterative Demarcation

Veldanda

Krishnamurthy

et al. 2022

IEEE Access

View full text Add to dashboard Cite

This paper proposes a novel feature-based on-line detection strategy, Removing Adversarial-Backdoors by Iterative Demarcation (RAID), for backdoor attacks. The proposed method is comprised of two parts: off-line training and on-line retraining. In the off-line training, a novelty detector and a shallow neural network are trained with clean validation data. During the on-line implementation, both models attempt to detect samples from the streaming data that differ from the validation data (i.e., flag likelypoisoned samples and possibly a few clean samples as false positives). An anomaly detector is used to purify the anomalous data by removing the clean samples. A binary support vector machine (SVM) is trained with the purified anomalous data and the clean validation data. RAID uses the SVM to detect poisoned inputs. To increase the accuracy as new anomalous data is being detected, the SVM is updated as well in real-time. It is shown that with updating, RAID can efficiently reduce the attack success rate while maintaining the classification accuracy against various types of backdoor attacks. The efficacy of RAID is compared against several state-of-the-art techniques. Additionally, it is shown that RAID only requires a small clean validation dataset to achieve such performance, and therefore provides a practical and efficient approach.

show abstract

“…Works of both [31] and [54] demonstrate that the open-set recognition can be benefited from generative features. It should be noted that open-set recognition is more challenging than novelty detection [29,32,36,37,30] because novelty detection only requires determining whether an observed image during inference belongs to one of the known classes. Self-Supervision.…”

Section: Related Workmentioning

confidence: 99%

Open-set Adversarial Defense with Clean-Adversarial Mutual Learning

Shao¹,

Perera²,

Yuen³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Open-set recognition and adversarial defense study two key aspects of deep learning that are vital for real-world deployment. The objective of open-set recognition is to identify samples from open-set classes during testing, while adversarial defense aims to robustify the network against images perturbed by imperceptible adversarial noise. This paper demonstrates that open-set recognition systems are vulnerable to adversarial samples. Furthermore, this paper shows that adversarial defense mechanisms trained on known classes are unable to generalize well to open-set samples. Motivated by these observations, we emphasize the necessity of an Open-Set Adversarial Defense (OSAD) mechanism. This paper proposes an Open-Set Defense Network with Clean-Adversarial Mutual Learning (OSDN-CAML) as a solution to the OSAD problem. The proposed network designs an encoder with dual-attentive feature-denoising layers coupled with a classifier to learn a noise-free latent feature representation, which adaptively removes adversarial noise guided by channel and spatial-wise attentive filters. Several techniques are exploited to learn a noise-free and informative latent feature space with the aim of improving the performance of adversarial defense and open-set recognition. First, we incorporate a decoder to

show abstract

Multiple Class Novelty Detection Under Data Distribution Shift

Cited by 16 publications

References 32 publications

Weakly Supervised Learning for Textbook Question Answering

Weakly Supervised Learning for Textbook Question Answering

A Feature-Based On-Line Detector to Remove Adversarial-Backdoors by Iterative Demarcation

Open-set Adversarial Defense with Clean-Adversarial Mutual Learning

Contact Info

Product

Resources

About