Exploiting unlabeled data in ensemble methods

Bennett, Kristin P.; Demiriz, Ayhan; Maclin, Richard

doi:10.1145/775047.775090

Cited by 121 publications

(82 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The pseudolabeled data is combined with the original labeled data and the classifier is trained again. The classifier we use is Discrete AdaBoost [4]. We will use this method as a baseline method in the experiments.…”

Section: Frame-level Labelsmentioning

confidence: 99%

See 1 more Smart Citation

VideoCut: Removing Irrelevant Frames by Discovering the Object of Interest

Liu

Hua

Chen

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We propose a novel method for removing irrelevant frames from a video given user-provided frame-level labeling for a very small number of frames. We first hypothesize a number of candidate areas which possibly contain the object of interest, and then figure out which area(s) truly contain the object of interest. Our method enjoys several favorable properties. First, compared to approaches where a single descriptor is used to describe a whole frame, each area's feature descriptor has the chance of genuinely describing the object of interest, hence it is less affected by background clutter. Second, by considering the temporal continuity of a video instead of treating the frames as independent, we can hypothesize the location of the candidate areas more accurately. Third, by infusing prior knowledge into the topic-motion model, we can precisely follow the trajectory of the object of interest. This allows us to largely reduce the number of candidate areas and hence reduce the chance of overfitting the data during learning. We demonstrate the effectiveness of the method by comparing it to several other semi-supervised learning approaches on challenging video clips.

show abstract

Section: Frame-level Labelsmentioning

confidence: 99%

“…We will use this method as a baseline method in the experiments. This kind of self-training [14] procedure has been used extensively in different domains [10][17] and achieved top results in the NIPS competition [4].…”

Section: Frame-level Labelsmentioning

confidence: 99%

VideoCut: Removing Irrelevant Frames by Discovering the Object of Interest

Liu

Hua

Chen

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…The learning process is a type of ensemble learning [9][10][11]. It involves multiple classifiers which label the unlabeled data to update and improve each other [12].…”

Section: Introductionmentioning

confidence: 99%

Disagreement-Based Multi-system Tracking

Wang

et al. 2013

Computer Vision - ACCV 2012 Workshops

View full text Add to dashboard Cite

Abstract. In this paper, we tackle the tracking problem from a fusion angle and propose a disagreement-based approach. While most existing fusion-based tracking algorithms work on different features or parts, our approach can be built on top of nearly any existing tracking systems by exploiting their disagreements. In contrast to assuming multi-view features or different training samples, we utilize existing well-developed tracking algorithms, which themselves demonstrate intrinsic variations due to their design differences. We present encouraging experimental results as well as theoretical justification of our approach. On a set of benchmark videos, large improvements (20% ∼ 40%) over the state-ofthe-art techniques have been observed.

show abstract

“…D´Alch´e Buc et al [2] showed that the error function can be extended and applied in the case of semi-supervised learning. K.P.Bennett [3] further proposed the ASSEMBLE algorithm, which works with any cost-sensitive base learner and at the same time has a simple, adaptive step-size rule. Although semi-supervised ensemble methods perform well in the empirical tests on both two-class and multi-class problems, they fail on complex dataset [2], where the labeled data will really bring crucial information that can't be obtained in unlabeled data such as the case in video dataset.…”

Section: Introductionmentioning

confidence: 99%

“…the unlabeled data can be assimilated into margin cost sensitive ensemble algorithms etc. [3]). Moreover, it takes advantages of active learning to accelerating the converging speed of the learning process.…”

Section: Introductionmentioning

confidence: 99%

Video Annotation by Active Learning and Semi-Supervised Ensembling

Yan

Hua

et al. 2006

2006 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

Supervised and semi-supervised learning are frequently applied methods to annotate videos by mapping low-level features into semantic concepts. Due to the large semantic gap, the main constraint of these methods is that the information contained in a limited-size labeled dataset can hardly represent the distributions of the semantic concepts. In this paper, we propose a novel semi-automatic video annotation framework, active learning with semi-supervised ensembling, which tries to tackle the disadvantages of current video annotation solutions. Firstly the initial training set is constructed based on distribution analysis of the entire video dataset. And then an active learning scheme is combined into a semi-supervised ensembling framework, which selects the samples to maximize the margin of the ensemble classifier based on both labeled and unlabeled data. Experimental results show that the proposed method performs superior to general semi-supervised learning algorithms and typical active learning algorithms in terms of annotation accuracy and stability.

show abstract

Exploiting unlabeled data in ensemble methods

Cited by 121 publications

References 12 publications

VideoCut: Removing Irrelevant Frames by Discovering the Object of Interest

VideoCut: Removing Irrelevant Frames by Discovering the Object of Interest

Disagreement-Based Multi-system Tracking

Video Annotation by Active Learning and Semi-Supervised Ensembling

Contact Info

Product

Resources

About