- talks - lists.cs.princeton.edu

FW: generals stuff
by Melissa Lawson 22 Apr '09

22 Apr '09

Peng Jiang will present his research seminar/general exam on Tuesday April 28 at 10AM in Room 402. The members of his committee are: Mona Singh (advisor), Olga Troyanskaya, and Tom Funkhouser. Everyone is invited to attend his talk and those faculty wishing to remain for the oral exam following are welcome to do so. His abstract and reading list follow below. ------------------------------------ Abstract: Large-scale biological networks are being determined at an increasing pace. Computational analyses of the resulting data provide new opportunities for revealing cellular organization and uncovering protein function and pathways. Clustering is one of the most widely used computational approaches for analyzing these types of biological data, and can reveal important modular structures. Previous network clustering algorithms in bio-informatics community work well on midsized networks such as, for example, the yeast protein interaction network. However, by recent computational techniques for data integration, we are starting to get large biological networks for numerous organisms. In these networks, there might be tens thousands of nodes and millions or even hundreds millions of interactions in a single graph. In such cases, previous biological clustering algorithms will be impractically slow or entirely unapplicable. In our work, we present a graph clustering algorithm with time complexity O(Vlog(V)+E) and space complexity O(E), where V and E are the number of vertices edges in the graph. By testing on several large genomic networks, our algorithm successfully clustered all of them, whereas most previous approaches failed. In the cases where previous approaches were able to cluster these networks, our algorithm was hundreds of times faster. We validated our approach with respect to its ability to uncover biologically relevant clusters and show we get state-of-the-art performance. Our fast algorithm enables us to perform new types of analysis. For example, we clustered a recent human context specific network data set over 230 different conditions, resulting in 230 networks which are nearly 300GB in total. We were able to uncover clusters specific to individual networks, putatively corresponding to context-specific modules, a type of analysis which was not possible prior to our work. Readlist: Book: [1] Thomas HC, Charles EL, Ronald LR, Clifford S. (2001) Introduction to algorithm, 2nd edition. MIT press Papers: [1] Broh´ee S., van Helden J. (2006) Evaluation of clustering algorithms for protein-protein interaction networks, BMC Bioinformatics, 7, 488. [2] Enright A.J., Van Dongen S., Ouzounis C.A. (2002) An efficient algorithm for largescale detection of protein families, Nucleic Acids Research, 30(7), 1575-1584. [3] Altaf-Ul-Amin M, Shinbo Y, Mihara K, Kurokawa K, Kanaya S. (2006) Development and implementation of an algorithm for detection of protein complexes in large interaction networks, BMC Bioinformatics, 7, 207. [4] Jensen LJ, Kuhn M, Stark M, Chaffron S, Creevey C, Muller J, Doerks T, Julien P, Roth A, Simonovic M, Bork P, von Mering C. (2009) STRING 8a global view on proteins and their functional interactions in 630 organisms, Nucleic Acids Res, 37, D412-416. [5] Bobby-Joe Breitkreutz, Chris Stark, Teresa Reguly, Lorrie Boucher, Ashton Breitkreutz, Michael Livstone, Rose Oughtred, Daniel Lackner, Jurg Bhler, Valerie Wood, Kara Dolinski and Mike Tyers. (2008) The BioGRID Interaction Database: 2008 Update, Nucleic Acids Res, 36, D637-640. [6] Curtis Huttenhower, Erin M. Haley3, Matthew A. Hibbs, Vanessa Dumeaux, Daniel R. Barrett, Hilary A. Coller, Olga G. Troyanskaya. (2009) Exploring the human genome with functional maps, Genome Research, 4(9), e1000165. [7] Fredman ML, Tarjan RE. (1987) Fibonacci heaps and their uses in improved network optimization algorithms, Journal of the ACM, 34(3), 596-615. [8] Rhee SY,Wood V, Dolinski K, Draghici S. (2008) Use and misuse of the gene ontology annotations, Nat Rev Genet, 9(7), 509-515. [9] Lord PW, Stevens RD, Brass A, Goble CA. (2003) Investigating semantic similarity measures across the Gene Ontology: the relationship between sequence and annotation, Bioinformatics, 19(10), 1275-1283.

1 0

4:30 p.m. Roger Johnston on physical and cyber security
by Andrew W. Appel 21 Apr '09

21 Apr '09

1 0

3pm Tue Apr 28 talk on "sustainable networks and systems for the developing world" in CS 402
by Jennifer Rexford 21 Apr '09

21 Apr '09

Speaker: Lakshmi Subramanian, NYU Title: Sustainable Networks and Systems for the Developing World Date/time: 3-4pm Tuesday April 28 Location: room 402 in the Computer Science building Abstract: The Internet and the World Wide Web have largely remained urban phenomena and a significant fraction of the developing world, especially in rural and underdeveloped regions, remains disconnected from the rest of the world. In this talk, I will briefly elaborate on our current efforts in addressing three broad challenges that can help in bridging the digital divide: (a) developing low-cost high-bandwidth connectivity solutions; (b) extending the Web to developing regions; (c) building new mobile applications for enhancing healthcare and financial services. This talk will specifically focus on the challenge of low-cost connectivity solutions for developing regions. I will describe our experiences in developing and deploying WiFi-based Long Distance (WiLD) Networks, a low-cost connectivity solution which provides the ability to achieve nearly 6-10 Mbps over 100s of kilometers for roughly $1000/link. WiLD networks have been deployed in 10 developing countries and have been used to provide telemedicine services in Aravind Eye hospitals in India to 50,000 patients/year in rural areas which is currently being expanded to cover 500,000 patients/year. We are currently extending WiLD networks to develop WiFi based Rural Extensions (WiRE), a low-cost alternative to cellular networks for rural settings. Bio Prof. Lakshminarayanan Subramanian is an Assistant Professor in the Courant Institute of Mathematical Sciences at NYU. His research interests are in the space of networks, distributed systems, security and technologies for developing regions. He received his MS and PhD from UC Berkeley and B.Tech from IIT-Madras.

1 0

DeCoro time correction!
by Melissa Lawson 20 Apr '09

20 Apr '09

Please correct start time for Chris DeCoro's FPO to 1:30PM, Monday April 27. Sorry for the confusion!

1 0

C DeCoro FPO
by Melissa Lawson 20 Apr '09

20 Apr '09

Chris DeCoro will present his FPO on Monday April 27 at 2PM in Room 402. The members of his committee are: Szymon Rusinkiewicz, advisor; Tom Funkhouser and Tim Weyrich (Univ College London), readers; Fei-Fei Li and Brian Kernighan, nonreaders. A copy of his thesis, "Rendering Filters for Controlling Detail and Creating Effects", is on display in Room 310. Everyone is invited to attend his talk.

1 0

P Calamia FPO
by Melissa Lawson 20 Apr '09

20 Apr '09

Paul Calamia will present his FPO on Monday April 27 at 10AM in Room 402. The members of his committee are: Tom Funkhouser, advisor; Perry Cook and Peter Svensson (NUST), readers; Adam Finkelstein and Szymon Rusinkiewicz, nonreaders. A copy of his thesis, "Advances in Edge-Diffraction Modeling for Virtual-Acoustic Simulations" is on display in Room 310. Everyone is invited to attend his talk.

1 0

FW: Michael Littman, Rutgers, TODAY, 4:25 PM
by Ginny Hogan 20 Apr '09

20 Apr '09

The Princeton-Yahoo! machine learning speaker series continues TODAY with Michael Littman from Rutgers University. Michael is the director of the Rutgers Laboratory for Real-Life Reinforcement Learning (RL^3) and his research in machine learning examines algorithms for decision making under uncertainty. Michael will be giving a talk TODAY, at 4:25pm, in CS 105 (small auditorium). The title and abstract for his talk are included below and can also be found on the website for the Princeton-Yahoo! speaker series: http://ai.cs.princeton.edu/pmwiki/pmwiki.php?n=Public.Talks. --- Efficiently Learning to Behave Efficiently Michael Littman Rutgers University The field of reinforcement learning is concerned with the problem of learning efficient behavior from experience. In real life applications, gathering this experience is time-consuming and possibly costly, so it is critical to derive algorithms that can learn effective behavior with bounds on the experience necessary to do so. This talk presents our successful efforts to create such algorithms within a novel learning-theory framework we call "KWIK" for "knows what it knows". I'll summarize the framework, our algorithms, their formal validations, and their empirical evaluations in robotic and videogame testbeds.

1 0

M Sun general exam
by Melissa Lawson 17 Apr '09

17 Apr '09

Min Sun will present his research seminar/general exam on Wednesday April 22 at 3:30PM in Room 402. The members of his committee are; Fei-Fei Li (advisor), David Blei, Adam Finkelstein. Everyone is invited to attend his talk and those faculty wishing to remain for the oral exam following are welcome to do so. His abstract and reading list follow below. ------------------------------------------------- We propose a novel probabilistic framework for learning visual models of 3D object categories by combining appearance information and geometric constraints. Objects are represented as a coherent ensemble of parts that are consistent under 3D viewpoint transformations. Each part is a collection of salient image features. A generative framework is used for learning a model that captures the relative position of parts within each of the discretized viewpoints. Contrary to most of the existing mixture of viewpoints models, our model establishes explicit correspondences of parts across different viewpoints of the object class. Given a new image, detection and classiﬁcation are achieved by determining the position and viewpoint of the model that maximize recognition scores of the candidate objects. Our approach is among the ﬁrst to propose a generative probabilistic framework for 3D object categorization. We test our algorithm on the detection task and the viewpoint classiﬁcation task by using “car” category from both the Savarese et al. 2007 and PASCAL VOC 2006 datasets. We show promising results in both the detection and viewpoint classiﬁcation tasks on these two challenging datasets. ======= Reading list: • Textbooks [1] C. M. Bishop. Pattern Recognition and Machine Learning (Information Science and Statistics). Springer, August 2006. Chapter 1, 2, 8, and 10. [2] D. A. Forsyth and J. Ponce. Computer Vision: A Modern Approach. Prentice Hall, August 2002. • Papers [1] D. M. Blei. Variational methods for the dirichlet process. In In Proceedings of the 21st International Conference on Machine Learning, 2004. [2] P. F. Felzenszwalb and D. P. Huttenlocher. Pictorial structures for object recognition. IJCV, 61:2005, 2005. [3] D. Hoeim, C. Rother, and J. Winn. 3d layoutcrf for multi-view object class recognition and segmentation. 2007. In Proc. In IEEE Conference on Computer Vision and Pattern Recognition. [4] A. Kushal, C. Schmid, , and J. Ponce. Flexible object models for category-level 3d object recognition. 2007. Flexible object models for category-level 3d object recognition. In Proc. In IEEE Conf. on Comp. Vis. and Patt. Recogn. [5] B. Leibe, A. Leonardis, and B. Schiele. Combined object categorization and segmentation with an implicit shape model. In In ECCV workshop on statistical learning in computer vision, pages 17–32, 2004. [6] J. Liebelt, C. Schmid, and K. Schertler. Viewpointindependent object class detection using 3d feature maps. Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on, pages 1–8, June 2008. [7] S. Savarese and L. Fei-Fei. 3d generic object categorization, localization and pose estimation. 2007. IEEE Intern. Conf. in Computer Vision (ICCV). [8] E. B. Sudderth, A. Torralba, W. T. Freeman, and A. S. Willsky. Describing visual scenes using transformed dirichlet processes. In In NIPS, pages 1297–1304. MIT Press, 2005. [9] A. Thomas, V. Ferrari, B. Leibe, T. Tuytelaars, B. Schiele, and L. V. Goo. Towards multi-view object class detection. In Proc. In IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 1589-1596. [10] P. Yan, D. Khan, and M. Shah. 3d model based object class detection in an arbitrary view. ICCV, 2007.

1 0

S Gaw FPO
by Melissa Lawson 01 Apr '09

01 Apr '09

Shirley Gaw will present her FPO on Wednesday April 8 at 10:30AM in 306 Sherrerd Hall (note room). The members of her committee are; Ed Felten, advisor: Paul Dourish, UC Irvine and Brian Kernighan, readers; Olga Troyanskaya and Perry Cook, nonreaders. A copy of her thesis, "Ideals and Reality: Adopting Secure Technologies and Developing Secure Habits to Prevent Message Disclosure", is on display in Room 310. Everyone is invited to attend her talk.

1 0

3pm talk today on content distribution and network evolution (room 402 in CS building)
by Jennifer Rexford 30 Mar '09

30 Mar '09

Speaker: Kobus van der Merwe, AT&T Lab--Research Title: It's the Content, Stupid. How Content is driving Network Evolution. Date/time: 3:00-4:00pm on Monday March 30 Location: room 402 in the CS building Abstract: Security and robustness, operational complexity and dealing with demand growth are arguably the most significant challenges facing large scale networks. In this talk I will show how Internet growth is driven by significant increases in content related traffic. Efficiently dealing with the distribution of this content presents a significant challenge to service providers. I will present our work on load-aware anycast which offers a cross-layer approach to content distribution (considering both content and network layers) . This work is currently being readied for trial in the AT&T network and directly builds on our earlier work on Closed-loop Route Control where route selection is informed by network intelligence. As such the work presents a successful case study on the transition from research concept to production. However, this work also serves to illustrate how difficult it is to make this transition, or indeed more generally, how difficult it is to effect any change in a provider network. In the second part of my talk I will present our recent work on a platform called ShadowNet. With ShadowNet we aim to address this fundamental challenge of network change by introducing an operational trial/test network which "sits between" a lab testbed and the production network. Specifically, each ShadowNet node is composed of a collection of carrier-grade equipment (routers, switches and servers). Each node is connected to both the Internet and other ShadowNet nodes via a (virtual) backbone. ShadowNet exploits the strong separation between logical functionality and physical realization that if offered by modern equipment. As such ShadowNet is a sharable, programmable and composable infrastructure, that allows multiple technology and service trials to be executed in parallel in a realistic operational setting. I will describe the ShadowNet control framework, our prototype implementation and deployment plans. Bio: Kobus Van der Merwe is a Principal Member of Technical Staff in the Internet and Networking Systems Research Center at AT&T Labs - Research, where he is fortunate to be doing networking systems research that are both forward looking and sometimes find application in the AT&T network. His current research interests include network management, control and operation, network security and content distribution.

2 1