%0 Generic %1 steeg2017unsupervised %A Steeg, Greg Ver %D 2017 %K corex fulltext gregversteeg mach shouldread softwarec statistics %T Unsupervised Learning via Total Correlation Explanation %U http://arxiv.org/abs/1706.08984 %X Learning by children and animals occurs effortlessly and largely without obvious supervision. Successes in automating supervised learning have not translated to the more ambiguous realm of unsupervised learning where goals and labels are not provided. Barlow (1961) suggested that the signal that brains leverage for unsupervised learning is dependence, or redundancy, in the sensory environment. Dependence can be characterized using the information-theoretic multivariate mutual information measure called total correlation. The principle of Total Cor-relation Ex-planation (CorEx) is to learn representations of data that "explain" as much dependence in the data as possible. We review some manifestations of this principle along with successes in unsupervised learning problems across diverse domains including human behavior, biology, and language.

@misc{steeg2017unsupervised, abstract = {Learning by children and animals occurs effortlessly and largely without obvious supervision. Successes in automating supervised learning have not translated to the more ambiguous realm of unsupervised learning where goals and labels are not provided. Barlow (1961) suggested that the signal that brains leverage for unsupervised learning is dependence, or redundancy, in the sensory environment. Dependence can be characterized using the information-theoretic multivariate mutual information measure called total correlation. The principle of Total Cor-relation Ex-planation (CorEx) is to learn representations of data that "explain" as much dependence in the data as possible. We review some manifestations of this principle along with successes in unsupervised learning problems across diverse domains including human behavior, biology, and language.}, added-at = {2018-10-04T20:10:14.000+0200}, author = {Steeg, Greg Ver}, biburl = {https://www.bibsonomy.org/bibtex/2cfe39b123dc8a8a8bc167a2b44910fa7/marcsaric}, description = {[1706.08984] Unsupervised Learning via Total Correlation Explanation}, interhash = {348961a62b619e72392718e1176024fb}, intrahash = {cfe39b123dc8a8a8bc167a2b44910fa7}, keywords = {corex fulltext gregversteeg mach shouldread softwarec statistics}, note = {cite arxiv:1706.08984Comment: Invited contribution for IJCAI 2017 Early Career Spotlight. 5 pages, 1 figure}, timestamp = {2018-10-04T20:10:14.000+0200}, title = {Unsupervised Learning via Total Correlation Explanation}, url = {http://arxiv.org/abs/1706.08984}, year = 2017 }