Fixed-Point Model For Structured Labeling

Quannan Li, Jingdong Wang, David Wipf, Zhuowen Tu
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(1):214-221, 2013.

Abstract

In this paper, we propose a simple but effective solution to the structured labeling problem: a fixed-point model. Recently, layered models with sequential classifiers/regressors have gained an increasing amount of interests for structural prediction. Here, we design an algorithm with a new perspective on layered models; we aim to find a fixed-point function with the structured labels being both the output and the input. Our approach alleviates the burden in learning multiple/different classifiers in different layers. We devise a training strategy for our method and provide justifications for the fixed-point function to be a contraction mapping. The learned function captures rich contextual information and is easy to train and test. On several widely used benchmark datasets, the proposed method observes significant improvement in both performance and efficiency over many state-of-the-art algorithms.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-li13b, title = {Fixed-Point Model For Structured Labeling}, author = {Quannan Li and Jingdong Wang and David Wipf and Zhuowen Tu}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {214--221}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/li13b.pdf}, url = {http://proceedings.mlr.press/v28/li13b.html}, abstract = {In this paper, we propose a simple but effective solution to the structured labeling problem: a fixed-point model. Recently, layered models with sequential classifiers/regressors have gained an increasing amount of interests for structural prediction. Here, we design an algorithm with a new perspective on layered models; we aim to find a fixed-point function with the structured labels being both the output and the input. Our approach alleviates the burden in learning multiple/different classifiers in different layers. We devise a training strategy for our method and provide justifications for the fixed-point function to be a contraction mapping. The learned function captures rich contextual information and is easy to train and test. On several widely used benchmark datasets, the proposed method observes significant improvement in both performance and efficiency over many state-of-the-art algorithms.} }
Endnote
%0 Conference Paper %T Fixed-Point Model For Structured Labeling %A Quannan Li %A Jingdong Wang %A David Wipf %A Zhuowen Tu %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-li13b %I PMLR %J Proceedings of Machine Learning Research %P 214--221 %U http://proceedings.mlr.press %V 28 %N 1 %W PMLR %X In this paper, we propose a simple but effective solution to the structured labeling problem: a fixed-point model. Recently, layered models with sequential classifiers/regressors have gained an increasing amount of interests for structural prediction. Here, we design an algorithm with a new perspective on layered models; we aim to find a fixed-point function with the structured labels being both the output and the input. Our approach alleviates the burden in learning multiple/different classifiers in different layers. We devise a training strategy for our method and provide justifications for the fixed-point function to be a contraction mapping. The learned function captures rich contextual information and is easy to train and test. On several widely used benchmark datasets, the proposed method observes significant improvement in both performance and efficiency over many state-of-the-art algorithms.
RIS
TY - CPAPER TI - Fixed-Point Model For Structured Labeling AU - Quannan Li AU - Jingdong Wang AU - David Wipf AU - Zhuowen Tu BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-li13b PB - PMLR SP - 214 DP - PMLR EP - 221 L1 - http://proceedings.mlr.press/v28/li13b.pdf UR - http://proceedings.mlr.press/v28/li13b.html AB - In this paper, we propose a simple but effective solution to the structured labeling problem: a fixed-point model. Recently, layered models with sequential classifiers/regressors have gained an increasing amount of interests for structural prediction. Here, we design an algorithm with a new perspective on layered models; we aim to find a fixed-point function with the structured labels being both the output and the input. Our approach alleviates the burden in learning multiple/different classifiers in different layers. We devise a training strategy for our method and provide justifications for the fixed-point function to be a contraction mapping. The learned function captures rich contextual information and is easy to train and test. On several widely used benchmark datasets, the proposed method observes significant improvement in both performance and efficiency over many state-of-the-art algorithms. ER -
APA
Li, Q., Wang, J., Wipf, D. & Tu, Z.. (2013). Fixed-Point Model For Structured Labeling. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(1):214-221

Related Material