UCLA Department of Statistics Speakers Series
1:00 PM-2:00 PM, Tuesday, January 31, 2012
5233 Math Sciences Bldg.
Kathrin Plath
Biological Chemistry
UCLA
Mechanisms of Reprogramming to Pluripotency
Ectopic expression of the four transcription factors Oct4, Sox2, cMyc, and Klf4 is sufficient to confer a pluripotent state to the human or murine somatic cell genome, generating induced pluripotent stem (iPS) cells. iPS cell lines are morphologically and functionally very similar to embryonic stem (ES) cells. Like ES cells, iPS cells can be induced to differentiate along lineages representative of the three embryonic germ layers. These results open the door to the generation of patient-specific stem cell lines with unlimited differentiation potential to study and potentially ameliorate human disease. We are interested in understanding the mechanisms underlying transcription factor- induced reprogramming to the iPS cell state.
UCLA Department of Statistics Speakers Series
3:00 PM-4:00 PM, Tuesday, January 31, 2012
3656 Geology Bldg.
Weng Kee Wong
Department of Biostatistics
UCLA
Nature-Inspired Metaheuristic Algorithms for Generating Optimal Experimental Designs
Particle swarm optimization (PSO) is a relatively new, simple and powerful way to search for an optimal solution. The method works quite magically and frequently finds the optimal solution or a nearly optimal solution quickly. There is virtually no assumption required for the method to perform well and the user only needs to input a few easy to work with tuning parameters.
I first review methods for finding optimal designs and present an overview of recent advances in the field with applications in the biomedical sciences. Using nonlinear regression models, I will demonstrate that once a model and an optimality criterion are specified, PSO can generate many types of optimal designs efficiently, including minimax optimal designs where effective algorithms to find such designs have remained stubbornly elusive until now.
This is joint work with Drs. Shin-Perng Chang and Weichung Wang of National Taiwan University and Dr. Ray-Bing Chen of National Cheng Kung University in Taiwan.
Project Cascade Among Fast Company's 22 Best Infographics for 2011 |
 Infographics can help struggling news organizations exploit the nuances of social media to reach more readers. Here, data-viz wiz Jer Thorp, UCLA prof Mark Hansen, and staff data scientist Jake Porway developed a way for the New York Times to visualize how stories spread across Twitter--info the paper can then use to program tweets that grow their audience (and, hopefully, their ad dollars). Find out more about the Cascade project at |
|
Teaching Computers To See |
 Statistics professor Song-Chu Zhu studies artificial intelligence to help computers understand one of the most challenging skills of all. Read more about Song-Chun's work in this UCLA College Report... |
|
Journal of Statistical Software Ranked Among "Rising Stars" Among Journals Two Years in a Row |
Thomson-Reuters, who also have the Journal of Statistical Software listed in computer science, have (for the second time) crowned the journal a "Rising Star", which means it is the fastest growing journal in computer science (in terms of their impact statistics). It would be nice if the journal could be listed in Statistics, but both Elsevier and Thomson are not really aware of the fact that Statistics exists (except as a branch of Mathematics). |
|
Jan de Leeuw on "History and Theory of Nonlinear Principal Component Analysis" |
A talk was given by Jan de Leeuw, Feb 11 on the occasion of the CARME 2011 organized by the Applied Mathematic Department Agrocampus Ouest. Title: History and Theory of Nonlinear Principal Component Analysis Abstract: Multiple Correspondence Analysis (MCA) is discussed as a form of Nonlinear Principal Component Analysis (NLPCA). It is compared with other forms of NLPCA that have been proposed over the years: Shepard-Kruskal-Breiman-Friedman-Gi PCA with optimal scaling, aspect analysis of correlations, Guttman's MSA, Logit/Probit PCA of binary data, and Logistic Homogeneity Analysis. Watch the presentation on YouTube at http://www.youtube.com/watch?v=V-hFORcBj44 |
|
UCLA Statistics @ Facebook |
The department now has its own Facebook page. The link is: http://www.facebook.com/uclastat Visit us at Facebook and become a friend. |
|
UCLA Statistics @ Twitter |
In addition to having a Facebook presence, the department has a Twitter presence at: http://twitter.com/#uclastat Visit us at Twitter and follow us there. |
|
Project Cascade |
Some of the most exciting work taking place in The New York Times building is being done on the 28th floor, in the paper's Research and Development Lab. For the past several months, the R&D Lab has been working, quietly, on a time-based representation of how the Times' news content is being shared in Twitter's social space. Its name: Project Cascade. Superficially, it's a data visualization, but it's actually a tool that could, ever so slightly, change the way we think about online engagement. It's the product of a collaboration among Mark Hansen, the UCLA stats professor who spent a spring 2010 sabbatical working at the Times as what Zimbalist calls the paper's "futurist-in-residence" along with Jer Thorp (data artist in residence) and Jake Porway (data scientist). "What it attempts to do," Zimbalist says, "is dimensionalize and make really physical and tangible the way that news is shared." Project Cascade — it's a working title — lets you visualize a cascade as a comprehensive unit; it also allows viewers to zoom in on particular events to see key points in how a story's sharing activity has unfolded. Its side view demonstrates a (relatively) simple timeline of a story's activity; its radar view demonstrates the separation between conversations; and its 3D view essentially marries the two, representing individual threads of conversation in a sharper way. Complex cascades can also be pruned to show only the most influential tweets (what Thorp calls the "backbone" of the conversation), and the users who posted them. The Project Cascade tool can also, intriguingly, help its users learn much more about the social elements of online news distribution. "In effect, what we're trying to do is think about notions of influence that are based not just on the social graph, but instead how that graph gets exercised," Hansen says. Or, in much more basic terms: to understand the audience — not as a passive monolith, but as a series of active individuals who happen also to be nodes in a network. The ability to comprehend communities will be crucial for organizations of all kinds going forward, news orgs first among those. Of the communities that form online, "do they represent different opinion groups?" Thorp says. "Do they represent different geographic groups? Do they represent groups that we've never identified before? Those are some questions that we can ask." Excerpt from N.Y. Times article "The New York Times' R&D Lab Has Built a Tool that Explores the Life Stories Take in the Social Space" by Megan Garber Read more: http://bit.ly/hmWM4Q |
|
Ian Fellows: 2011 John M. Chambers Statistical Software Award Winner |
It gives us great pleasure to announce that Ian Fellows' "Deducer's Plot Builder: A General GUI for Statistical Graphics" has been selected as the winner of the 2011 Chambers Award. The prize consists of a plaque, a check for $1000, and the opportunity to receive the award in person at the 2011 Joint Statistical Meetings in Miami Beach, FL. The meetings run from July 30 to August 4, 2011, and the prize is currently scheduled to be awarded on Monday evening, August 1, at the Joint Statistical Computing and Statistical Graphics mixer. As part of the award, the Statistical Computing section will pay for your registration in the meetings and travel expenses for transportation and hotel to attend the meetings (up to reasonable amount). Congratulations Ian. |
|
JSM 2011 Presentations |
This year's Joint Statistical Meeting will be held in Miami Beach, Florida, July 30 through August 4. Several UCLA Statistics department members will be participating, including: Peter Bentler, Mine Cetinkaya, Nicolas Christou, Ivo Dinov, Siavash Jalal, Judea Pearl, Colin Rundel, Kekona Sorenson, Brigid Wilson, Dave Zes, Zhongyang Zhang. See listing below for their participation. For more information about JSM 2011, please visit http://www.amstat.org/meetings/jsm/2011/. UCLA Statistics Participants
|
|
Song-Chun Zhu Elevated to IEEE Fellow |
Following the IEEE Board of Directors meeting in November 2010, Dr. Song-Chun Zhu has been elevated to IEEE Fellow for "contributions to statistical modeling, learning and inference in computer vision". This is one of the Association's most prestigious honors and takes effect January 1, 2011. |
|
MURI Grant Awarded to Song-Chun Zhu |
Song-Chun Zhu has been awarded a grant by the Office of Naval Research to conduct research under the office's Multidisciplinary University Research Initiative (MURI). The $7.5m grant was granted to do research under the topic "Reasoning for Image Understanding in Uncertain Environments", specifically: "Knowledge Representation, Reasoning and Learning for Understanding Scenes and Events". The P.I., Dr. Zhu, is joined by international leaders in human and computer vision, AI and reasoning, cognitive science, and statistical machine learning to develop automated computer vision systems that are capable of the following:
Dr. Zhu's MURI web site is located at: http://muri.stat.ucla.edu. The work is supported by the Office of Naval Research grant N000141010933 The program manager is Behzad Kamgar-Parsi. |
|
The New Distributome Project |
In October 2010, the National Science Foundation (NSF) funds the multi-institutional collaborative research project entitled: Distributome — An Interactive Web-based Resource for Probability Distributions. The principal investigators of the Distributome project are Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH). Link to URL: http://www.Distributome.org |
|
Welcome
Why Statistics?
The world is becoming more and more quantitative. Many professions depend on numerical measurements to make decisions
in the face of uncertainty. Statisticians use quantitative abilities, statistical knowledge, and communication skills
to work on many challenging problems.
Want to Learn More?
The Department of Statistics at UCLA coordinates undergraduate and graduate statistics teaching and research within the College of Letters and Sciences. We teach a large number of undergraduates and we have a substantial graduate program. Our research and teaching have a strong emphasis on computational and applied statistics. We have an active consulting center for both on-campus and off-campus clients.
You can find the Department’s catalog on-line. We offer nearly one hundred courses; an undergraduate B.S. degree and Minor; and M.S. and Ph.D. degrees in Statistics.
Want to Help?
Donations and contributions of any amount to the Department of Statistics are greatly appreciated. Generous giving from the community, alumni, and parents helps the department to maintain high standards of computational technology and continue with its rapid growth. Thank you for your consideration. For more information please visit our webpage for online giving.
