Crowdsourcing Feature Discovery via Adaptively Chosen Comparisons
作者： Adam Tauman Kalai , Kamalika Chaudhuri , James Y. Zou
DOI:
关键词: Crowdsourcing 、 Computer science 、 Feature (computer vision) 、 Data set 、 Machine learning 、 Feature learning 、 Artificial intelligence 、 Simple (abstract algebra) 、 Adaptive algorithm 、 Data mining
摘要: We introduce an unsupervised approach to efficiently discover the underlying features in a data set via crowdsourcing. Our queries ask crowd members articulate feature common two out of three displayed examples. In addition, we provide binary labels for these discovered on remaining The triples are chosen adaptively based previously set. This is motivated by formal framework elicitation that and analyze this paper. natural models features, hierarchical independent, show simple adaptive algorithm recovers all with less labor than any nonadaptive algorithm. savings as result automatically avoiding redundant or synonyms. Experimental results validate theoretical findings usefulness approach.
arxiv.org
harvard.edu
crowdml.cc 参考文章(14)
Tamara L. Berg, Alexander C. Berg, Jonathan Shih, Automatic attribute discovery and characterization from noisy web data european conference on computer vision. pp. 663- 676 ,(2010) , 10.1007/978-3-642-15549-9_48
Antti Ukkonen, Hannes Heikinheimo, The Crowd-Median Algorithm. national conference on artificial intelligence. ,(2013)
Adam Kalai, Serge J. Belongie, Omer Tamuz, Omer Tamuz, Ohad Shamir, Ce Liu, Adaptively Learning the Crowd Kernel international conference on machine learning. pp. 673- 680 ,(2011)
Amos Tversky, None, Features of Similarity Psychological Review. ,vol. 84, pp. 327- 352 ,(1977) , 10.1037/0033-295X.84.4.327
G. Patterson, J. Hays, SUN attribute database: Discovering, annotating, and recognizing scene attributes computer vision and pattern recognition. pp. 2751- 2758 ,(2012) , 10.1109/CVPR.2012.6247998
Josiah Wang, Katja Markert, Mark Everingham, Learning Models for Object Recognition from Natural Language Descriptions british machine vision conference. pp. 1- 11 ,(2009) , 10.5244/C.23.2
Devi Parikh, Kristen Grauman, Interactively building a discriminative vocabulary of nameable attributes computer vision and pattern recognition. pp. 1681- 1688 ,(2011) , 10.1109/CVPR.2011.5995451
Justin Cheng, Michael S. Bernstein, Flock: Hybrid Crowd-Machine Learning Classifiers conference on computer supported cooperative work. pp. 600- 611 ,(2015) , 10.1145/2675133.2675214
Ali Farhadi, Ian Endres, Derek Hoiem, David Forsyth, Describing objects by their attributes computer vision and pattern recognition. pp. 1778- 1785 ,(2009) , 10.1109/CVPR.2009.5206772
Alexis Battle, Andrew Y. Ng, Honglak Lee, Rajat Raina, Efficient sparse coding algorithms neural information processing systems. ,vol. 19, pp. 801- 808 ,(2006)