Learning with dataset bias in latent subcategory models

Stamos, Dimitris; Martelli, Samuele; Nabi, Moin; McDonald, Andrew; Murino, Vittorio; Pontil, Massimiliano

doi:10.1109/cvpr.2015.7298988

Cited by 7 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This ensures that the performance measured off-line on iCubWorld can be expected to generalize well when the system is deployed on the actual robot. Note that this aspect of iCubWorld is extremely relevant since visual biases make it typically difficult to generalize performances across different datasets and applications, as already well known from previous work (Pinto et al 2008;Torralba and Efros 2011;Khosla et al 2012;Hoffman et al 2013;Rodner et al 2013;Model and Shamir 2015;Stamos et al 2015;Tommasi et al 2015) and also shown empirically in Sec. 4.1 of this paper.…”

Section: The Icubworld Transformations Datasetmentioning

confidence: 70%

Are we done with object recognition? The iCub robot’s perspective

Pasquale

Ciliberto

Odone

et al. 2019

Robotics and Autonomous Systems

View full text Add to dashboard Cite

We report on an extensive study of the benefits and limitations of current deep learning approaches to object recognition in robot vision scenarios, introducing a novel dataset used for our investigation. To avoid the biases in currently available datasets, we consider a natural human-robot interaction setting to design a data-acquisition protocol for visual object recognition on the iCub humanoid robot. Analyzing the performance of off-the-shelf models trained off-line on largescale image retrieval datasets, we show the necessity for knowledge transfer. We evaluate different ways in which this last step can be done, and identify the major bottlenecks affecting robotic scenarios. By studying both object categorization and identification problems, we highlight key differences between object recognition in robotics applications and in image retrieval tasks, for which the considered deep learning approaches have been originally designed. In a nutshell, our results confirm the remarkable improvements yield by deep learning in this setting, while pointing to specific open challenges that need be addressed for seamless deployment in robotics.

show abstract

Section: The Icubworld Transformations Datasetmentioning

confidence: 70%

Are we done with object recognition? The iCub robot’s perspective

Pasquale

Ciliberto

Odone

et al. 2019

Robotics and Autonomous Systems

View full text Add to dashboard Cite

show abstract

“…Usually, the researchers apply feature selection technique on a dataset and train the machine learning models over the selected features. Apparently, in this way, the machine learning models perform good on one dataset but do not perform well when tested on different dataset [29]. Therefore, we propose a universal feature set that can efficiently support the machine algorithms for better detection of the botnet attacks irrespective of a dataset.…”

Section: Features Selectionmentioning

confidence: 99%

Towards a Universal Features Set for IoT Botnet Attacks Detection

Hussain¹,

Fayyaz

Shah³

et al. 2020

Preprint

View full text Add to dashboard Cite

The security pitfalls of IoT devices make it easy for the attackers to exploit the IoT devices and make them a part of a botnet. Once hundreds of thousands of IoT devices are compromised and become the part of a botnet, the attackers use this botnet to launch the large and complex distributed denial of service (DDoS) attacks which take down the target websites or services and make them unable to respond the legitimate users. So far, many botnet detection techniques have been proposed but their performance is limited to a specific dataset on which they are trained. This is because the features used to train a machine learning model on one botnet dataset, do not perform well on other datasets due to the diversity of attack patterns. Therefore, in this paper, we propose a universal features set to better detect the botnet attacks regardless of the underlying dataset. The proposed features set manifest preeminent results for detecting the botnet attacks when tested the trained machine learning models over three different botnet attack datasets.

show abstract

“…Higher-level Representation. Most of the existing work tackle this problem by learning domain invariant and compact representation from multiple source domains [16,58,73,74,111,132,162,163,207]. For example, Khosla et al [111] explicitly model the bias of each source domain and try to estimate the weights for the unbiased data by removing the source domain biases.…”

Section: Unavailable Target Datamentioning

confidence: 99%

“…Multiple sources can be generalised to a target task, referred to as domain generalization [16,58,73,74,111,132,162,163,207] without the need of any target data. Domain generalization is of practical significance, but less addressed in the previous research.…”

Section: Transfer Learning From Multiple Sourcesmentioning

confidence: 99%

Recent Advances in Transfer Learning for Cross-Dataset Visual Recognition

Zhang

Ogunbona

et al. 2019

ACM Comput. Surv.

View full text Add to dashboard Cite

This paper takes a problem-oriented perspective and presents a comprehensive review of transfer learning methods, both shallow and deep, for cross-dataset visual recognition. Specifically, it categorises the cross-dataset recognition into seventeen problems based on a set of carefully chosen data and label attributes. Such a problem-oriented taxonomy has allowed us to examine how different transfer learning approaches tackle each problem and how well each problem has been researched to date. The comprehensive problem-oriented review of the advances in transfer learning with respect to the problem has not only revealed the challenges in transfer learning for visual recognition, but also the problems (e.g. eight of the seventeen problems) that have been scarcely studied.This survey not only presents an up-to-date technical review for researchers, but also a systematic approach and a reference for a machine learning practitioner to categorise a real problem and to look up for a possible solution accordingly.

show abstract

Learning with dataset bias in latent subcategory models

Cited by 7 publications

References 23 publications

Are we done with object recognition? The iCub robot’s perspective

Are we done with object recognition? The iCub robot’s perspective

Towards a Universal Features Set for IoT Botnet Attacks Detection

Recent Advances in Transfer Learning for Cross-Dataset Visual Recognition

Contact Info

Product

Resources

About