Multi-View Learning in the Presence of View Disagreement

By C. M. Christoudias, Raquel Urtasun, Trevor Darrell. 2008

146 citations recorded by the source corpus.

Traditional multi-view learning approaches suffer in the presence of view disagreement, i.e., when samples in each view do not belong to the same class due to view corruption, occlusion or other noise processes. In this paper we present a multi-view learning approach that uses a conditional entropy criterion to detect view disagreement. Once detected, samples with view disagreement are filtered and standard multi-view learning methods can be successfully applied to the remaining samples. Experimental evaluation on synthetic and audio-visual databases demonstrates that the detection and filtering of view disagreement considerably increases the performance of traditional multi-view learning approaches.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Multiagent Learning: Basics, Challenges, and Prospects (2012, neutral): “…and develops learning algorithms inspired by natural evolution, using operators like selection, mutation, crossover, and others; multiview learning (Christoudias, Urtasun, and Darrell 2008), that is machine-learning methods that use redundant views of the same input data; multistrategy learning…”
A Survey on Multi-view Learning (2013, methodology): “Christoudias et al. (2009b) presented a Bayesian localized approach for combining different feature representations with Gaussian processes that learns a local weighting over each view.”
Semi-supervised Domain Adaptation with Instance Constraints (2013, methodology): “Instance similarity constraints have been used in multiview learning [4, 11, 21], canonical-correlation analysis [15], and as constraints between domains when available [24].”
Multimodal Machine Learning: A Survey and Taxonomy (2017, background): “It has also been extended to deal with disagreement between modalities, by filtering out unreliable samples [43].”
Inductive multi-task learning with multiple view data (2012, background): “Multi-view semi-supervised learning has attracted significant research interest in recent years [1, 5, 11, 14, 19, 23, 24].”
Constrained Semi-supervised Learning in the Presence of Unanticipated Classes (2016, background): “Prior work [33] has also studied correlation between the performance of multi-view methods and view agreement.”
Learning to Rank Using Privileged Information (2013, background): “Alternatively, one can use the training data to learn a mapping from the image to the privileged modality and use this predictor to fill in the values missing at test time [5].”
Multi-view prediction of protein function (2011, unknown): “Multi-view learning has been applied to natural-language processing [6], document categorization [10, 17] and signal processing [7].”