Multi-Class Active Learning by Uncertainty Sampling with Diversity Maximization

Yang, Yi; Ma, Zhiliang; Nie, Feiping; Chang, Xiaojun; Hauptmann, Alexander G.

doi:10.1007/s11263-014-0781-x

Cited by 391 publications

(207 citation statements)

References 29 publications

Supporting

Mentioning

206

Contrasting

Unclassified

Order By: Relevance

“…Since it is impractical to use human data for real-world retargeting tasks, our method aims to generate AGPs to mimick human gaze shifting, where experimental results show a high degree of consistency (over 90%) of our AGP and real human gaze shifting path. Yang et al [50] proposed a semi-supervised batch mode multiclass learning algorithm for visual concept recognition, which exploits the whole active pool to evaluate the data uncertainty. Yang et al [51] built upon the assumption that different-related tasks share common structures.…”

Section: Hmentioning

confidence: 99%

Perceptually Guided Photo Retargeting

Xia

Zhang

Hong

et al. 2017

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Abstract-We propose perceptually guided photo retargeting, which shrinks a photo by simulating a human's process of sequentially perceiving visually/semantically important regions in a photo. In particular, we first project the local features (graphlets in this paper) onto a semantic space, wherein visual cues such as global spatial layout and rough geometric context are exploited. Thereafter, a sparsity-constrained learning algorithm is derived to select semantically representative graphlets of a photo, and the selecting process can be interpreted by a path which simulates how a human actively perceives semantics in a photo. Furthermore, we learn the prior distribution of such active graphlet paths (AGPs) from training photos that are marked as esthetically pleasing by multiple users. The learned priors enforce the corresponding AGP of a retargeted photo to be maximally similar to those from the training photos. On top of the retargeting model, we further design an online learning scheme to incrementally update the model with new photos that are esthetically pleasing. The online update module makes the algorithm less dependent on the number and contents of the initial training data. Experimental results show that: 1) the proposed AGP is over 90% consistent with human gaze shifting path, as verified by the eye-tracking data, and 2) the retargeting algorithm outperforms its competitors significantly, as AGP is more indicative of photo esthetics than conventional saliency maps.

show abstract

Section: Hmentioning

confidence: 99%

Perceptually Guided Photo Retargeting

Xia

Zhang

Hong

et al. 2017

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…For instance, Guo et al proposed an active learning method which selects a set of instances by maximizing the likelihood of labeled and selected instances, while minimizing the uncertainty of unlabeled instances [15]. For multi-class visual concept recognition tasks, the approach proposed in [16] adopted the entropy theory to evaluate the uncertainty of unlabeled instances and simultaneously incorporates diversity criterion to make the informative selection as diverse as possible. Long et al [17] proposed a novel multi-annotator Guassian process model to deal with multiclass visual recognition in the collaborative active learning framework and fully explored the trade-off between exploitation and exploration criteria.…”

Section: Related Workmentioning

confidence: 99%

Multi-label active learning by model guided distribution matching

Gao

Huang

Chen

2016

Front. Comput. Sci.

View full text Add to dashboard Cite

Multi-label learning is an effective framework for learning with objects that have multiple semantic labels, and has been successfully applied into many real-world tasks. In contrast with traditional single-label learning, the cost of labeling a multi-label example is rather high, thus it becomes an important task to train an effective multi-label learning model with as few labeled examples as possible. Active learning, which actively selects the most valuable data to query their labels, is the most important approach to reduce labeling cost. In this paper, we propose a novel approach MADM for batch mode multi-label active learning. On one hand, MADM exploits representativeness and diversity in both the feature and label space by matching the distribution between labeled and unlabeled data. On the other hand, it tends to query predicted positive instances, which are expected to be more informative than negative ones. Experiments on benchmark datasets demonstrate that the proposed approach can reduce the labeling cost significantly.

show abstract

“…Different from most existing AL methods that can only query labels for selected instances at the class level, their approach established a semantic framework that predicted scene labels based on a latent object-based image representation, and was capable of querying labels at two different levels-the sceneclass level and the latent object-class level. Yang et al [46] proposed a semi-supervised batch mode multi-class AL algorithm for visual concept recognition. Chang et al [7] proposed a novel convex, semi-supervised multi-label feature selection algorithm applicable to large-scale datasets.…”

Section: Active Learningmentioning

confidence: 99%

Active domain adaptation with noisy labels for multimedia analysis

et al. 2015

View full text Add to dashboard Cite

Supervised learning methods require sufficient labeled examples to learn a good model for classification or regression. However, available labeled data are insufficient in many applications. Active learning (AL) and domain adaptation (DA) are two strategies to minimize the required amount of labeled data for model training. AL requires the domain expert to label a small number of highly informative examples to facilitate classification, while DA involves tuning the source domain knowledge for classification on the target domain. In this paper, we demonstrate how AL can efficiently minimize the required amount of labeled data for DA. Since the source and target domains usually have different distributions, it is possible that the domain expert may not have sufficient knowledge to answer World Wide Web each query correctly. We exploit our active DA framework to handle incorrect labels provided by domain experts. Experiments with multimedia data demonstrate the efficiency of our proposed framework for active DA with noisy labels.

show abstract

Multi-Class Active Learning by Uncertainty Sampling with Diversity Maximization

Cited by 391 publications

References 29 publications

Perceptually Guided Photo Retargeting

Perceptually Guided Photo Retargeting

Multi-label active learning by model guided distribution matching

Active domain adaptation with noisy labels for multimedia analysis

Contact Info

Product

Resources

About