- talks - lists.cs.princeton.edu

TODAY 4:30pm: CITP/PIIRS GSR public lecture: "Internet Resilience: Global Network, Global Risks" with Jennifer Rexford & Nick Feamster, Wed, Nov 30, 4:30pm, CS 105
by Nicole E. Wagenblast 30 Nov '16

30 Nov '16

This afternoon at 4:30pm in Computer Science 105 (Small Auditorium), the Center for Information Technology Policy (CITP) and the PIIRS Global Systemic Risk research community are co-hosting a public lecture and discussion on "Internet Resilience: Global Network, Global Risks" with Jennifer Rexford and Nick Feamster. Jen is Chair of the Computer Science Department at Princeton and Nick is Professor of Computer Science and Acting Director of CITP. In the study of systemic risk, the Internet is an informative case study as it is our most recently created system, consciously designed and constructed with the goal of avoiding fragility and failure. Internet resilience has emerged as an increasingly critical area of study, as more and more of our systems—finance, infrastructure, transportation, health care, agriculture, energy & electricity, among others—are interconnected with and interdependent upon the Internet. In addition to the impact of the Internet on these macro systems that support our lives, we are increasingly affected by Internet security on an individual level, as many of our consumer products—such as smartphones, refrigerators, thermostats, and cars—are becoming connected to and reliant upon the Internet in the emerging "Internet of Things" (IoT). Jen and Nick will discuss architecture and design of the Internet, how it operates, security risks, and instances of attacks and failures. One topic of discussion will be security threats posed by IoT addressed in recent CITP research ( news and report ). Brief bios for Jen and Nick and an abstract for their talk are below. The lecture will be held this afternoon, Wednesday, November 30th, at 4:30pm in Computer Science 105 (Small Auditorium), and will be followed by Q&A and discussion. The talk is open to the public—please share this invitation with members of your department and all those you think may be interested. ***************** Abstract: The Internet has rapidly become a critical part of the world's communications infrastructure. Yet this global “network of networks” is remarkably vulnerable to cyberattacks, physical damage, and manipulation of the user experience through censorship, wiretapping, and propaganda. In this public discussion, we discuss how the Internet consists of tens of thousands of networks owned by different institutions in different countries, with different economic and political goals. We discuss how early decisions in the design of the Internet led to both tremendous innovation and the substantial security and privacy risks. The architecture of the Internet is a case study in the design and construction of resilient networks and systems. However, as more of our global systems—and our lives and livelihoods—become increasingly interdependent on the Internet, the potential fragility of the Internet is an increasing cause for concern and academic study. Jennifer Rexford is the Gordon Y.S. Wu Professor of Engineering, and Chair of the Computer Science Department. Before joining Princeton in 2005, she worked for eight years at AT&T Labs—Research. She is a member of the American Academy of Arts and Sciences and the National Academy of Engineering and was the 2004 winner of ACM's Grace Murray Hopper Award for outstanding young computer professional. She is co-author of the book Web Protocols and Practice (2001). She received her B.S.E. in electrical engineering from Princeton, and her Ph.D. in electrical engineering and computer science from the University of Michigan. Nick Feamster is Professor of Computer Science and Acting Director of the Center for Information Technology Policy (CITP) at Princeton. His research focuses on networked computer systems, with a strong emphasis on network operations, network architecture and protocol design, high performance wired and wireless networks, and anti-censorship techniques and systems. The collective goal of these research areas is to help network operators to manage their networks better and enable users of these networks to experience high availability and end-to-end performance. He received both his B.S. in electrical engineering and computer science and his Ph.D. in computer science from MIT. *************

1 0

Colloquium Speaker: Alexis Battle, Tuesday, today, 12:30pm
by Nicole E. Wagenblast 29 Nov '16

29 Nov '16

Colloquium Speaker Prof. Alexis Battle, Johns Hopkins University Tuesday, November 29, 2016 - 12:30pm Computer Science 105 Predicting the impact of rare regulatory variation The enormous increase in availability of full human genomic sequences presents great opportunity for understanding the impact of rare genetic variants. Based on current knowledge, however, we are still limited in our ability to interpret or predict regulatory consequences of rare variants. Association methods are inherently limited for variants observed in only one or a few individuals. Diverse genomic annotations such as growing resources of epigenetic data and have been shown to be informative regarding regulatory elements, but are only moderately predictive of impact for individual variants. The availability of personal RNA-seq and other cellular measurements for the same individuals with genome sequencing offers a new avenue for integrated methods for prioritizing rare functional variants. By considering informative genomic annotations along with molecular phenotyping, we are able to identify the regulatory impact of rare genetic variants largely excluded from previous analyses. We have evaluated the impact of rare regulatory variants using whole genome sequences and corresponding RNA-sequence in 44 different tissues samples from the GTEx project and report. We demonstrate that rare variants in promoter elements, conserved regions, and variants that affect splicing are associated with extreme changes in gene expression across multiple tissues. Additionally, we have developed a Bayesian machine learning approach that integrates whole genome sequencing with RNA-seq data from the same individual, leveraging gene expression levels along with diverse genomic annotations and performing joint inference to identify likely functional rare regulatory variants. We have applied this model to the GTEx data to prioritize the most likely functional rare regulatory variants for each individual. We demonstrate that integrative models perform better than predictions from DNA-sequencing or RNA-sequencing alone. Our probabilistic model of rare regulatory genetic variants offers potential for identifying deleterious regulatory variants from individual genomes. Alexis Battle’s research focuses on unraveling the impact of genetics on the human health, using machine learning and probabilistic methods to analyze large scale genomic data. She is interested in applications to personal genomics, genetics of gene expression, and gene networks in disease, leveraging diverse data to infer more comprehensive models of genetic effects on the cell. She earned her Ph.D. and Masters in Computer Science in 2014 from Stanford University in 2014, where she also received her Bachelors in Symbolic Systems (2003). Alexis also spent several years in industry as a member of the technical staff at Google. Prior to joining Hopkins, Alexis spent a year as a postdoc with Jonathan Pritchard with HHMI and the Genetics Department at Stanford. She joined JHU in July 2014.

1 0

Katie Anna Wolf will present her Pre-FPO "Techniques and Tools for End-User Sonification Design and Audio Personalization" on Monday, December 5th, 2016 at 10am CS 302.
by Nicki Gotsis 28 Nov '16

28 Nov '16

Katie Anna Wolf will present her Pre-FPO on Monday, December 5th, 2016 at 10am CS 302. The members of her committee are as follows: Examiners: Rebecca Fiebrink (advisor); Dan Truman (Music), and Marshini Chetty. Readers: Adam Finkelstein and Mike Nees (Lafayette College) Everyone is invited to attend her talk. The talk title and abstract follow below: Title: Techniques and Tools for End-User Sonification Design and Audio Personalization Abstract: Modern daily life involves consuming increasing amounts of data from personalized applications and devices such as Twitter, the stock market and health trackers. As the amount and types of data increase, we need to continue to develop technology that presents the information in such a way that its meaning can be interpreted not only by scientists but by everyday people. Sonification, as a sister to visualization, is a way of representing data with sound. Due to physical impairments, high-dimensional data that is hard to visualize, or just a general desire to avoid staring at a screen, audio interfaces have been developed to take advantage of the communicative power of sound. However designing sonifications requires understanding of how the end users will use the system as well as knowledge in many domains including sound design, sonification design, and programming. In this talk I will present methods that give users who are non-experts in sound and sonification design the ability to customize their own sonifications using a graphical user interface. These methods include: (1) the analysis and modeling of environmental soundscapes (including animal vocalizations and sounds of the weather) which allow non-experts to talk about sounds in terms they know, (2) using related work by sonification experts and my own exploration of environmental sounds to design an automatic mapping technique that determines how the data should be represented by sounds, and (3) using an example-based approach that allows users to select example sound sets as an exemplar for how the sonification should sound. I will also discuss my ongoing work in developing tools that allow end users to control and personalize their auditory environment. Finally I will present a survey about people’s experience with audio personalization, and how the results may influence the design of tools that put the end user at the center of the sonification design process.

1 0

Colloquium Speaker: Alexis Battle, Tuesday, Nov 29, 12:30pm- reminder
by Nicole E. Wagenblast 28 Nov '16

28 Nov '16

Colloquium Speaker Prof. Alexis Battle, Johns Hopkins University Tuesday, November 29, 2016 - 12:30pm Computer Science 105 Predicting the impact of rare regulatory variation The enormous increase in availability of full human genomic sequences presents great opportunity for understanding the impact of rare genetic variants. Based on current knowledge, however, we are still limited in our ability to interpret or predict regulatory consequences of rare variants. Association methods are inherently limited for variants observed in only one or a few individuals. Diverse genomic annotations such as growing resources of epigenetic data and have been shown to be informative regarding regulatory elements, but are only moderately predictive of impact for individual variants. The availability of personal RNA-seq and other cellular measurements for the same individuals with genome sequencing offers a new avenue for integrated methods for prioritizing rare functional variants. By considering informative genomic annotations along with molecular phenotyping, we are able to identify the regulatory impact of rare genetic variants largely excluded from previous analyses. We have evaluated the impact of rare regulatory variants using whole genome sequences and corresponding RNA-sequence in 44 different tissues samples from the GTEx project and report. We demonstrate that rare variants in promoter elements, conserved regions, and variants that affect splicing are associated with extreme changes in gene expression across multiple tissues. Additionally, we have developed a Bayesian machine learning approach that integrates whole genome sequencing with RNA-seq data from the same individual, leveraging gene expression levels along with diverse genomic annotations and performing joint inference to identify likely functional rare regulatory variants. We have applied this model to the GTEx data to prioritize the most likely functional rare regulatory variants for each individual. We demonstrate that integrative models perform better than predictions from DNA-sequencing or RNA-sequencing alone. Our probabilistic model of rare regulatory genetic variants offers potential for identifying deleterious regulatory variants from individual genomes. Alexis Battle’s research focuses on unraveling the impact of genetics on the human health, using machine learning and probabilistic methods to analyze large scale genomic data. She is interested in applications to personal genomics, genetics of gene expression, and gene networks in disease, leveraging diverse data to infer more comprehensive models of genetic effects on the cell. She earned her Ph.D. and Masters in Computer Science in 2014 from Stanford University in 2014, where she also received her Bachelors in Symbolic Systems (2003). Alexis also spent several years in industry as a member of the technical staff at Google. Prior to joining Hopkins, Alexis spent a year as a postdoc with Jonathan Pritchard with HHMI and the Genetics Department at Stanford. She joined JHU in July 2014.

1 0

Fwd: Princeton Optimization Seminar, Bernard Chazelle, Thursday, Dec. 1, 4:30 PM, Sherrerd 101
by Nicole E. Wagenblast 28 Nov '16

28 Nov '16

----- Princeton Optimization Seminar ----- DATE: TODAY Thursday, December 1, 2016 TIME : 4:30PM LOCATION: Sherrerd Hall 101 SPEAKER: Bernard Chazelle , Princeton University TITLE: The Mathematics of Natural Algorithms Abstract: I will review some of the recent techniques we've used in our study of natural algorithms. These include Dirichlet series for matrix products, mean-field approximations in opinion dynamics, graph sequence grammars, anti-mixing methods for iterated learning, and tools for renormalizing network-based dynamical systems. The talk will be self-contained and non-technical. Bio: Bernard Chazelle is Eugene Higgins Professor of Computer Science at Princeton University, where he has been on the faculty since 1986. His current research focuses on the “algorithmic nature” of living systems. A professor at the Collège de France in Paris in recent years as well as a member of the Institute for Advanced Study in Princeton, he received his PhD in computer science from Yale University in 1980. The author of the book, "The Discrepancy Method," he is a fellow of the American Academy of Arts and Sciences, the European Academy of Sciences, the Association for Computing Machinery, and the recipients of three Best-Paper awards from SIAM. *** If you would like to subscribe to the mailing list of the optimization seminar series, please visit the link below or send an email with SUBSCRIBE opt-seminar in the body to listserv(a)lists.princeton.edu . https://lists.princeton.edu/cgi-bin/wa?SUBED1=opt-seminar&A=1

1 0

Colloquium Speaker: Alexis Battle, Tuesday, Nov 29, 12:30pm
by Nicole E. Wagenblast 22 Nov '16

22 Nov '16

Colloquium Speaker Prof. Alexis Battle, Johns Hopkins University Tuesday, November 29, 2016 - 12:30pm Computer Science 105 Predicting the impact of rare regulatory variation The enormous increase in availability of full human genomic sequences presents great opportunity for understanding the impact of rare genetic variants. Based on current knowledge, however, we are still limited in our ability to interpret or predict regulatory consequences of rare variants. Association methods are inherently limited for variants observed in only one or a few individuals. Diverse genomic annotations such as growing resources of epigenetic data and have been shown to be informative regarding regulatory elements, but are only moderately predictive of impact for individual variants. The availability of personal RNA-seq and other cellular measurements for the same individuals with genome sequencing offers a new avenue for integrated methods for prioritizing rare functional variants. By considering informative genomic annotations along with molecular phenotyping, we are able to identify the regulatory impact of rare genetic variants largely excluded from previous analyses. We have evaluated the impact of rare regulatory variants using whole genome sequences and corresponding RNA-sequence in 44 different tissues samples from the GTEx project and report. We demonstrate that rare variants in promoter elements, conserved regions, and variants that affect splicing are associated with extreme changes in gene expression across multiple tissues. Additionally, we have developed a Bayesian machine learning approach that integrates whole genome sequencing with RNA-seq data from the same individual, leveraging gene expression levels along with diverse genomic annotations and performing joint inference to identify likely functional rare regulatory variants. We have applied this model to the GTEx data to prioritize the most likely functional rare regulatory variants for each individual. We demonstrate that integrative models perform better than predictions from DNA-sequencing or RNA-sequencing alone. Our probabilistic model of rare regulatory genetic variants offers potential for identifying deleterious regulatory variants from individual genomes. Alexis Battle’s research focuses on unraveling the impact of genetics on the human health, using machine learning and probabilistic methods to analyze large scale genomic data. She is interested in applications to personal genomics, genetics of gene expression, and gene networks in disease, leveraging diverse data to infer more comprehensive models of genetic effects on the cell. She earned her Ph.D. and Masters in Computer Science in 2014 from Stanford University in 2014, where she also received her Bachelors in Symbolic Systems (2003). Alexis also spent several years in industry as a member of the technical staff at Google. Prior to joining Hopkins, Alexis spent a year as a postdoc with Jonathan Pritchard with HHMI and the Genetics Department at Stanford. She joined JHU in July 2014.

1 0

Colloquium Speaker Chinmay Kulkarni, Today- 12:30pm
by Nicole E. Wagenblast 17 Nov '16

17 Nov '16

Colloquium Speaker Chinmay Kulkarni, Carnegie Mellon University Thursday, November 17, 2016 - 12:30pm Computer Science 105 Structuring Peer Interactions for Learning Expertise at Scale Learning with peers helps students reflect, generalize knowledge and apply it more successfully to new problems. How can we scale successful peer learning from the controlled environment of the small classroom to the wild, massive scale of online classes? In his talk, Chinmay Kulkarni will introduce computational systems that structure peer learning at massive scale. These systems demonstrate how insights from educational theory can be distilled into interfaces that scale teaching to thousands of learners. For instance, he will describe how students using his PeerStudio obtain improvement-oriented feedback on open-ended work in just twenty minutes at any time of day, enabling them to revise and gain mastery. Similarly, students use the Talkabout system to engage in globe-spanning discussions that improve learning through diversity. This talk will be a reflection of the growing maturity of research in this area, from articulating design patterns, to randomized controlled experiments at scale, to studies of long-term effects on learners and learning. Finally, Kulkarni hopes to reflect on how the large scale and diversity of online classes can enrich learning at universities, and his ongoing efforts to prepare engineering students for tomorrow's grand challenges. Chinmay Kulkarni is an Assistant Professor in the Human-Computer Interaction Institute at Carnegie Mellon University. His research group investigates how new software and pedagogical systems can leverage peer processes at massive scale. Over the past three years, his work has been used by more than 100,000 learners online, in classes taught by more than twenty universities, including Harvard. In his idle time, Chinmay is either making coffee, drinking coffee, or clicking "View Source" on web pages he visits.

1 0

Colloquium Speaker Chinmay Kulkarni, Thurs. Nov 17th- 12:30pm- reminder
by Nicole E. Wagenblast 16 Nov '16

16 Nov '16

Colloquium Speaker Chinmay Kulkarni, Carnegie Mellon University Thursday, November 17, 2016 - 12:30pm Computer Science 105 Structuring Peer Interactions for Learning Expertise at Scale Learning with peers helps students reflect, generalize knowledge and apply it more successfully to new problems. How can we scale successful peer learning from the controlled environment of the small classroom to the wild, massive scale of online classes? In his talk, Chinmay Kulkarni will introduce computational systems that structure peer learning at massive scale. These systems demonstrate how insights from educational theory can be distilled into interfaces that scale teaching to thousands of learners. For instance, he will describe how students using his PeerStudio obtain improvement-oriented feedback on open-ended work in just twenty minutes at any time of day, enabling them to revise and gain mastery. Similarly, students use the Talkabout system to engage in globe-spanning discussions that improve learning through diversity. This talk will be a reflection of the growing maturity of research in this area, from articulating design patterns, to randomized controlled experiments at scale, to studies of long-term effects on learners and learning. Finally, Kulkarni hopes to reflect on how the large scale and diversity of online classes can enrich learning at universities, and his ongoing efforts to prepare engineering students for tomorrow's grand challenges. Chinmay Kulkarni is an Assistant Professor in the Human-Computer Interaction Institute at Carnegie Mellon University. His research group investigates how new software and pedagogical systems can leverage peer processes at massive scale. Over the past three years, his work has been used by more than 100,000 learners online, in classes taught by more than twenty universities, including Harvard. In his idle time, Chinmay is either making coffee, drinking coffee, or clicking "View Source" on web pages he visits.

1 0

Colloquium Speaker Chinmay Kulkarni, Thurs. Nov 17th- 12:30pm
by Nicole E. Wagenblast 11 Nov '16

11 Nov '16

Colloquium Speaker Chinmay Kulkarni, Carnegie Mellon University Thursday, November 17, 2016 - 12:30pm Computer Science 105 Structuring Peer Interactions for Learning Expertise at Scale Learning with peers helps students reflect, generalize knowledge and apply it more successfully to new problems. How can we scale successful peer learning from the controlled environment of the small classroom to the wild, massive scale of online classes? In his talk, Chinmay Kulkarni will introduce computational systems that structure peer learning at massive scale. These systems demonstrate how insights from educational theory can be distilled into interfaces that scale teaching to thousands of learners. For instance, he will describe how students using his PeerStudio obtain improvement-oriented feedback on open-ended work in just twenty minutes at any time of day, enabling them to revise and gain mastery. Similarly, students use the Talkabout system to engage in globe-spanning discussions that improve learning through diversity. This talk will be a reflection of the growing maturity of research in this area, from articulating design patterns, to randomized controlled experiments at scale, to studies of long-term effects on learners and learning. Finally, Kulkarni hopes to reflect on how the large scale and diversity of online classes can enrich learning at universities, and his ongoing efforts to prepare engineering students for tomorrow's grand challenges. Chinmay Kulkarni is an Assistant Professor in the Human-Computer Interaction Institute at Carnegie Mellon University. His research group investigates how new software and pedagogical systems can leverage peer processes at massive scale. Over the past three years, his work has been used by more than 100,000 learners online, in classes taught by more than twenty universities, including Harvard. In his idle time, Chinmay is either making coffee, drinking coffee, or clicking "View Source" on web pages he visits.

1 0

Colloquium Speaker Uri Keich, Thurs Nov 10, 12:30pm
by Nicole E. Wagenblast 10 Nov '16

10 Nov '16

Colloquium Speaker Prof. Uri Keich, University of Sydney Thursday, November 10, 12:30pm Computer Science 105 Controlling the rate of false discoveries in tandem mass spectrum identifications A typical shotgun proteomics experiment produces thousands of tandem mass spectra, each of which can be tentatively assigned a corresponding peptide by using a database search procedure that looks for a peptide-spectrum match (PSM) that optimizes the score assigned to a matched pair. Some of the resulting PSMs will be correct while others will be false, and we have no way to verify which is which. The statistical problem we face is of controlling the false discovery rate (FDR), or the expected proportion of false PSMs among all reported pairings. While there is a rich statistical literature on controlling the FDR in the multiple hypothesis testing context, controlling the FDR in the PSM context is mostly done through the "home grown" method called target-decoy competition (TDC). After a brief introduction to the problem of tandem mass spectrum identification we will explore the reasons why the mass spec community has been using this non-standard approach to controlling the FDR. We will then discuss how calibration can increase the number of correct discoveries and offer an alternative method for controlling the FDR in the presence of calibrated scores. We will conclude by arguing that our analysis extends to a more general setup than the mass spectrum identification problem. Joint work with Bill Noble (University of Washington)

1 0