#  Statistics Colloquium: Harrison Zhou (Yale University) 

 



####  calendar\_today Date and Time 

 **November 23, 2020** 

 10:30AM - 11:30AM EST 

####  pin\_drop Location 

 **Zoom - please contact emilie_campanelli@fas.harvard.edu for more information**  



 

 



 

###  Title:

 Global Convergence of EM?

###  Abstract:

 In this talk I will first discuss a recent joint work with Yihong Wu: [https://arxiv.org/abs/1908.10935](https://urldefense.proofpoint.com/v2/url?u=https-3A__arxiv.org_abs_1908.10935&d=DwMGaQ&c=WO-RGvefibhHBZq3fL85hQ&r=cSZTSObb9KOZYAv_RgNPYOFsxuPhRQql-BOH56RsWMSZzhUjq3GA4enOiQZiC44G&m=QSQA8HjsOn_KvadA8q95LUjrx6NUnRcQjSHhfbN-q8s&s=vwO4TeZAQJ_Niq2TLiBmE3xwvREMRGBZdbOR4DVVkSI&e=). We show that the randomly initialized EM algorithm for parameter estimation in the symmetric two-component Gaussian mixtures converges to the MLE in at most $\\sqrt(n)$ iterations with high probability. Then I will mention the limitations of that work and propose an extension to general Gaussian mixtures by overparameterization.



 

 



 

 See also:- [ Colloquia ](/event-type/colloquia)
 
 

 Share on:- [     Facebook ](#)
- [     Twitter ](#)
- [     Linkedin ](#)
 


 Save: [ Add to calendar calendar\_today ](https://statistics.fas.harvard.edu/node/1369590/event-feed.ics)  Copy link link