PPT-Unified Expectation Maximization

Rajhans Samdani Joint work with MingWei Chang Microsoft Research and Dan Roth University of Illinois at UrbanaChampaign Page 1 NAACL 2012 Montreal Weakly Supervised Learning in NLP. Hongning Wang. CS@UVa. Today’s lecture. k. -means clustering . A typical . partitional. . clustering . algorithm. Convergence property. Expectation Maximization algorithm. Gaussian mixture model. . Honglei. . Zhuang. , . Yihan. Sun, Jie Tang, Jialin Zhang, Xiaoming Sun. Influence Maximization. 0.6. 0.5. 0.1. 0.4. 0.6. 0.1. 0.8. 0.1. A. B. C. D. E. F. Probability . of . influence. Marketer Alice. Machine Learning. Last Time. Expectation Maximization. Gaussian Mixture Models. Today. EM Proof. Jensen’s Inequality. Clustering sequential data. EM over . HMMs. EM in any Graphical Model. Gibbs Sampling. Mixture Models and Expectation Maximization. Machine Learning. Last Time. Review of Supervised Learning. Clustering. K-means. Soft K-means. Today. Gaussian Mixture Models. Expectation Maximization. The Problem. Maximum Likelihood (ML) Model. Introduction. . Alternative Splicing. Simulation Setup: . human genome data (UCSC hg18) . UCSC database - 66, 803 . isoforms. 19, 372 genes, Single error-free reads: 60M of length 100bp. Zhizhuo. Zhang . Outline. Review of Mixture Model and EM algorithm. Importance Sampling. Re-sampling EM. Extending EM. Integrate Other Features. Result. Review Motif Finding: Mixture modeling. Given a dataset . Honglei. . Zhuang. , . Yihan. Sun, Jie Tang, Jialin Zhang, Xiaoming Sun. Influence Maximization. 0.6. 0.5. 0.1. 0.4. 0.6. 0.1. 0.8. 0.1. A. B. C. D. E. F. Probability . of . influence. Marketer Alice. Machine Learning. April 13, 2010. Last Time. Review of Supervised Learning. Clustering. K-means. Soft K-means. Today. A brief look at Homework 2. Gaussian Mixture Models. Expectation Maximization. The Problem. Machine . Learning . 10-601. , Fall . 2014. Bhavana. . Dalvi. Mishra. PhD student LTI, CMU. Slides are based . on materials . from . Prof. . Eric Xing, Prof. . . William Cohen and Prof. Andrew Ng. Xinran He . and David Kempe. University of Southern . California. {xinranhe, . dkempe. }@usc.edu. 08/15/2016. The adoption of new products . can . propagate between nodes . in the social network. 0.8. Pursue . Excellence . – Be the best. Expectation Cards. At Castle View we expect our students to conduct themselves in an exemplary manner at all times and to follow the school rules. . To help students with this, we have introduced an expectation card that all . 1. Matt Gormley. Lecture . 24. November 21, 2016. School of Computer Science. Readings:. 10-601B Introduction to Machine Learning. Reminders. Final . Exam. in-. class. . Wed. ., . Dec. . 7. 2. Outline. Winter 2012. Daniel Weld. Slides adapted from Carlos . Guestrin. , Dan Klein & Luke . Zettlemoyer. Machine Learning. 2. Supervised Learning. Parametric. Reinforcement Learning. Unsupervised Learning. B-Cell . Lymphoma (DLBCL) Patients, . 1983 . – 2014. results from . analysis of US SEER data. Ron . Dewar, . Registry and . Analytics,. Nova Scotia Health Authority (Canada). Nadia . Howlader. , Angela .