- talks - lists.cs.princeton.edu

Reminder: CS and CITP Virtual Lunch Seminar: Seny Kamara, tomorrow at 12:30pm
by Emily C. Lawrence 20 Apr '20

20 Apr '20

CS and CITP Virtual Lunch Seminar Speaker: Seny Kamara, Brown University Date: Tuesday, April 21 Time: 12:30-1:30pm Location: Webinar Host: Arvind Narayanan Event page: [ https://www.cs.princeton.edu/events/25947 | https://www.cs.princeton.edu/events/25947 ] Please click here to register for this webinar: [ https://princeton.zoom.us/webinar/register/WN_3P3w-nJ5Td6RdHBeL4j_Kg | https://princeton.zoom.us/webinar/register/WN_3P3w-nJ5Td6RdHBeL4j_Kg ] After registering, you will receive a confirmation email containing information about joining the webinar. This virtual talk is being co-sponsored by CITP and the Department of Computer Science. It is open only to Princeton University faculty, staff, and students. Title: Cryptography for Privacy and Policy Abstract: As our lives become increasingly digital and we produce more and more data, we are witnessing several conflicting trends. On one hand, these massive datasets are becoming more intrusive and privacy-sensitive and on the other they are becoming harder to protect. This is illustrated by the constant occurrence of data breaches in almost every industry and sector. One of the best tools at our disposal for enforcing data privacy and security is end-to-end encryption where only the data owner hold the encryption key. While end-to-end encryption is already deployed in messaging and video calling apps, its widespread adoption is severely limited because it breaks the utility of critical technologies like databases, cloud computing and machine learning. In this talk, I will describe my work on designing, analyzing and cryptanalyzing efficient algorithms and systems that operate on end-to-end encrypted data. I will discuss the advances my collaborators and I have made on these problems over the last fifteen years based on ideas from a variety of fields including cryptography, algorithms, data structures, complexity theory, machine learning and databases. In addition to the technological impact of these advances, I will also discuss their societal implications which range from (potentially) expanding Law Enforcement's "Going Dark" problem to enabling new policy tradeoffs. Bio: Seny Kamara is an Associate Professor of Computer Science at Brown University, where he directs the Encrypted Systems Lab and is affiliated with the Center for Human Rights and Humanitarian Studies. Prior to joining Brown, he was a Research Scientist at Microsoft Research in the Cryptography Research group. At Brown, he conducts research in cryptography with a focus on problems motivated by social and policy issues. In 2016, he was appointed by the National Academies of Sciences to study the impact of end-to-end encryption on law enforcement and intelligence agencies and in 2019 he testified to the U.S. House of Representative about the privacy and fairness implications of Big Data. He has received a Google Faculty Award and was named a Leadership Fellow by the Boston Global Forum for his work and commitment to global peace.

1 0

CS and CITP Virtual Lunch Seminar: Seny Kamara, Tuesday April 21
by Emily C. Lawrence 15 Apr '20

15 Apr '20

CS and CITP Virtual Lunch Seminar Speaker: Seny Kamara, Brown University Date: Tuesday, April 21 Time: 12:30-1:30pm Location: Webinar Host: Arvind Narayanan Event page: [ https://www.cs.princeton.edu/events/25947 | https://www.cs.princeton.edu/events/25947 ] Please click here to register for this webinar: [ https://princeton.zoom.us/webinar/register/WN_3P3w-nJ5Td6RdHBeL4j_Kg | https://princeton.zoom.us/webinar/register/WN_3P3w-nJ5Td6RdHBeL4j_Kg ] After registering, you will receive a confirmation email containing information about joining the webinar. This virtual talk is being co-sponsored by CITP and the Department of Computer Science. It is open only to Princeton University faculty, staff, and students. Title: Cryptography for Privacy and Policy Abstract: As our lives become increasingly digital and we produce more and more data, we are witnessing several conflicting trends. On one hand, these massive datasets are becoming more intrusive and privacy-sensitive and on the other they are becoming harder to protect. This is illustrated by the constant occurrence of data breaches in almost every industry and sector. One of the best tools at our disposal for enforcing data privacy and security is end-to-end encryption where only the data owner hold the encryption key. While end-to-end encryption is already deployed in messaging and video calling apps, its widespread adoption is severely limited because it breaks the utility of critical technologies like databases, cloud computing and machine learning. In this talk, I will describe my work on designing, analyzing and cryptanalyzing efficient algorithms and systems that operate on end-to-end encrypted data. I will discuss the advances my collaborators and I have made on these problems over the last fifteen years based on ideas from a variety of fields including cryptography, algorithms, data structures, complexity theory, machine learning and databases. In addition to the technological impact of these advances, I will also discuss their societal implications which range from (potentially) expanding Law Enforcement's "Going Dark" problem to enabling new policy tradeoffs. Bio: Seny Kamara is an Associate Professor of Computer Science at Brown University, where he directs the Encrypted Systems Lab and is affiliated with the Center for Human Rights and Humanitarian Studies. Prior to joining Brown, he was a Research Scientist at Microsoft Research in the Cryptography Research group. At Brown, he conducts research in cryptography with a focus on problems motivated by social and policy issues. In 2016, he was appointed by the National Academies of Sciences to study the impact of end-to-end encryption on law enforcement and intelligence agencies and in 2019 he testified to the U.S. House of Representative about the privacy and fairness implications of Big Data. He has received a Google Faculty Award and was named a Leadership Fellow by the Boston Global Forum for his work and commitment to global peace.

1 0

CITP Webinar: Ed Felten – COVID-19, Technology, Privacy and Civil Liberties, Thursday, April 16
by Emily C. Lawrence 14 Apr '20

14 Apr '20

CITP Webinar: Ed Felten – COVID-19, Technology, Privacy and Civil Liberties Date: Thursday, April 16, 2020 Time: 3:00 pm - 4:30 pm Event page: [ https://citp.princeton.edu/event/webinar-felten-covid/ | https://citp.princeton.edu/event/webinar-felten-covid/ ] Please click here to register for the webinar: [ https://princeton.zoom.us/webinar/register/WN_bAmNsJQsSYKSveh2_ef2FA | https://princeton.zoom.us/webinar/register/WN_bAmNsJQsSYKSveh2_ef2FA ] This event is open to the public. Many systems have been proposed for using technology to help individuals and public health officials better respond to the COVID-19 pandemic. This talk will analyze the major proposed uses of information technology in the public health response to COVID-19, including aggregate reporting, contact tracing via direct proximity detection or location history matching, and creation of disease status passports. The public health value of these approaches will be considered along with their privacy and civil liberties implications. For several approaches, broad public acceptance is a prerequisite for success, making careful privacy and civil liberties protection an important contributor to public health goals. Bio: Edward W. Felten is the Robert E. Kahn Professor of Computer Science and Public Affairs at Princeton University and the founding director of Princeton’s Center for Information Technology Policy. In 2011-12 he served as the first chief technologist for the Federal Trade Commission. His research interests include computer security and privacy, especially relating to media and consumer products; and technology law and policy. He has published about 80 papers in the research literature and two books. His research on topics such as web security, copyright and copy protection, and electronic voting has been covered extensively in the popular press. His weblog, at freedom-to-tinker.com, is widely read for its commentary on technology, law and policy. Professor Felten is a member of the National Academy of Engineering and the American Academy of Arts and Sciences, and is a fellow of the ACM. He has testified at House and Senate committee hearings on privacy, electronic voting and digital television. In 2004, Scientific American magazine named him to its list of 50 worldwide science and technology leaders.

1 0

Yichen Chen will present his FPO "On the complexity of Markov decision problems" on Friday, 4/3/2020 at 10:45am in B327 EQuad
by Nicki Mahler 01 Apr '20

01 Apr '20

Yichen Chen will present his FPO "On the complexity of Markov decision problems" on Friday, 4/3/2020 at 10:45am in B327 EQuad. The members of his committee are as follows: Mengdi Wang (Adviser); Readers: Yuxin Chen and Karthik Narasimhan; Examiners: Sanjeev Arora, Robert Schapire, and Mengdi Wang. A copy of his thesis is available upon request (please email [ mailto:ngotsis@cs.princeton.edu | ngotsis(a)cs.princeton.edu ] ). All are welcome to attend. The talk abstract follows below. The Markov decision problem (MDP) is one of the most basic models for sequential decision-making problems in a dynamic environment where outcomes are partly random. It models a stochastic control process in which a planner makes a sequence of decisions as the system evolves. The Markov decision problem provides a mathematical framework for dynamic programming, stochastic control, and reinforcement learning. In this thesis, we study the complexity of solving MDPs. In the first part of the thesis, we propose a class of stochastic primal-dual methods for solving MDPs. We formulate the core policy optimization problem of the MDP as a stochastic saddle point problem. By utilizing the value-policy duality structure, the algorithm samples state-action transitions and makes alternative updates to value and policy estimates. We prove that our algorithm is able to find the approximate optimal policies to Markov decision problems with small space and computational complexity. Using linear models to represent the value functions and the policies, our algorithm is capable of scaling to problems with infinite and continuous state spaces. In the second part of the thesis, we establish the computational complexity lower bounds for solving MDPs. We prove our results by modeling the MDP algorithms using branching programs and then characterizing the properties of these programs by quantum arguments. The analysis is also extended to study the complexity of solving two-player turn-based Markov games. Our results show that if we have a simulator that can sample according to the transition probability function in O(1), the lower bounds have reduced dependence on the state number. These results suggest a fundamental di↵erence between Markov decision problems with and without a simulator. We believe that our results provide a new piece of theoretical evidence for the success of simulation-based methods in solving MDPs and Markov games.

1 1

TOMORROW Wednesday, April 1 at 4:30pm: EE VIRTUAL SEMINAR - David Kohlbrenner
by Emily C. Lawrence 31 Mar '20

31 Mar '20

EE SEMINAR SERIES Speaker: David Kohlbrenner, UC Berkeley Title: Building Trusted Systems on Top of Leaky Abstractions Day: Wednesday, April 1, 2020 Time: 4:30 pm Zoom link: [ https://princeton.zoom.us/j/6092587282 | https://princeton.zoom.us/j/6092587282 ] Host: Prateek Mittal Event page: [ https://ee.princeton.edu/events/building-trusted-systems-top-leaky-abstract…: | https://ee.princeton.edu/events/building-trusted-systems-top-leaky-abstract…: ] Abstract: A modern application's safety and security relies on a staggeringly complex stack of abstractions, from software libraries and operating systems, to firmware and chip architectural choices. Often, performance-oriented design choices at a lower level can impact security in surprising ways, and have no clean fix. This talk will cover how I construct secure and deployable systems without needing to know the adversary's exact technique. To understand the threat posed by leaky abstractions, I first detail my work on a novel class of side-channel attacks. With this as motivation, I present my Fuzzyfox project for securing the Firefox web browser against all known and unknown timing attacks. Finally, I describe a new way to construct trusted systems leveraging both software and hardware in the Keystone Trusted Execution Environment Framework and future projects. Bio: David Kohlbrenner is a Postdoctoral Scholar working with Dawn Song at UC Berkeley. He previously received his PhD from UC San Diego where he was advised by Hovav Shacham. His research focuses on building deployable secure systems with both hardware and software. Adaptations of his research have been deployed in Firefox, Chrome and the Linux kernel. His adversarial research discovered major vulnerabilities in popular web browsers, resulting in extensive changes to Chrome, Safari, and Firefox. David also co-founded the San Diego-based embedded security company Somerset Recon in 2012.

1 0

Ruth Dannenfelser will present her FPO "Data-Driven Approaches and Systems to Interrogate Complex Disease" on Monday, 3/30/2020 at 1:00 pm (via Zoom)
by Nicki Mahler 30 Mar '20

30 Mar '20

Ruth Dannenfelser's FPO at 1pm today will now take place via Zoom: [ https://princeton.zoom.us/j/867970222 | https://princeton.zoom.us/j/867970222 ] -------------------------------------------------------- Ruth Dannenfelser will present her FPO "Data-Driven Approaches and Systems to Interrogate Complex Disease" on Monday, 3/30/2020 at 1:00 pm via Zoom [ https://princeton.zoom.us/j/867970222 | https://princeton.zoom.us/j/867970222 ] . The members of her committee are as follows: Olga Troyanskaya (adviser); Readers: Wendell Lim (UCSF) and Kai Li; Examiners: Yibin Kang (Mol Bio), Mona Singh, and Olga Troyanskaya. A copy of her thesis is available upon request. Please [ mailto:ngotsis@cs.princeton | email ngotsis(a)cs.princeton ] if you would like a copy of the thesis. Everyone is invited to attend her talk. The talk abstract follows below. Large-scale genomic studies now give more predictive power than ever, allowing us to profile the composition of tissues, study cellular functions, and understand organismal traits at an unprecedented level of detail. This is particularly important for studying heterogeneous diseases, such as cancer, where small patient-specific differences play critical roles in disease development and progression. As the these studies accumulate, it is increasingly important to develop methods to discover novel biology while considering tissue and cell type specificity, and develop systems to help make this data explosion easily manageable, accessible, and interpretable. Towards these goals, in this dissertation, we build off the wealth of publicly available data to examine the interplay between cancer and the immune system, then develop two query-based visualization systems that enable interactive data exploration for the wider biomedical community. The first part of this work presents two perspectives on cancer and the immune system. Using derived immune markers we found that estrogen receptor activity and genomic complexity are key factors driving changes in breast cancer lymphocytic infiltration. Our method enabled discoveries on existing samples even when this was not the original intent of the study, without the need for additional experiments. Next, we leveraged public expression data to further the development of targeted immunotherapeutics for solid tumors. Working closely with experimental collaborators, we developed a method to prioritize pairs of antigen targets that will help engineered T cells hone in on tumor targets while minimizing damage to healthy tissues. The second half covers how we can extract unbiased signals from large collections of biomedical data in the form of abstracts and repositories of transcriptomics data. We develop a method to obtain informative tissue-disease-gene relationships from abstracts and integrate them into a system that presents different snapshots of curated interactions. Next, we extend a gene expression search engine that simultaneously returns coexpressed genes and relevant datasets. Our extension expands the search space across the major model organisms and provides a new cross-organism exploration interface to help facilitate translational research. Both systems will help experimentalists leverage existing knowledge to better explain the larger implications of their specific findings.

1 0

Colloquium Speaker Ravi Netravali TODAY at 12:30pm
by Emily C. Lawrence 23 Mar '20

23 Mar '20

CS Colloquium Speaker Speaker: Ravi Netravali, University of California, Los Angeles Date: Monday, March 23 - 12:30pm Host: Karthik Narasimhan Location: Remote via Zoom Link: https://princeton.zoom.us/j/645162020 Event page: https://www.cs.princeton.edu/events/25925 ***Due to the developing situation surrounding the COVID-19 virus, this talk will be available for remote viewing. See below for details.*** Title: Improving Stack-Wide Resource Utilization for a Faster Mobile Web Abstract: Mobile web pages are integral to today's society, supporting critical services such as education, e-commerce, and social networking. Despite considerable academic and industrial research efforts, and major improvements over the past decade across the client-side web stack (i.e., networks, device CPUs, and browser engines), page load performance has plateaued and continues to fall short of user performance demands in practice. The consequences of this are far reaching: users abandon pages early, costing content providers billions of dollars in lost revenue; or pages are unusably slow, particularly in developing regions where web pages are often the sole gateway to the aforementioned services. In this talk, I will describe the origin of this performance plateau in the context of serialized page load tasks that preclude effective utilization of the underlying network and CPU resources. Then, I will describe two complementary optimizations that my students and I have developed to eliminate these inefficiencies throughout the page load process and cut mobile load times in half. Key to these optimizations are judicious applications of programming languages (e.g., symbolic execution) and machine learning (e.g., reinforcement learning) techniques that enable us to 1) discover optimization knobs that preserve application correctness, and 2) tune those knobs according to stack-wide signals from the network, device, page, and browser, without developer intervention. I will conclude by describing how these underlying techniques can motivate and address a range of future challenges in networked applications and distributed systems. Bio: Ravi Netravali is an Assistant Professor of Computer Science at UCLA. His research interests are broadly in computer systems and networking, with a recent focus on building practical systems to improve the performance and debugging of large-scale, distributed applications for both end users and developers. His research has been recognized with an NSF CAREER Award, a Google Faculty Research Award, an ACM SoCC Best Paper Award, and an IRTF Applied Networking Research Prize. Prior to joining UCLA, Netravali received a PhD in Computer Science from MIT in 2018. Zoom information: Topic: Ravi Netravali CS Seminar Time: Mar 23, 2020 12:00 PM Eastern Time (US and Canada) Join Zoom Meeting https://princeton.zoom.us/j/645162020 Meeting ID: 645 162 020 One tap mobile +13126266799,,645162020# US (Chicago) +16465588656,,645162020# US (New York) Dial by your location +1 312 626 6799 US (Chicago) +1 646 558 8656 US (New York) +1 253 215 8782 US +1 301 715 8592 US +1 346 248 7799 US (Houston) +1 669 900 6833 US (San Jose) Meeting ID: 645 162 020 Find your local number: https://princeton.zoom.us/u/avcvlf1F3 Join by SIP 645162020(a)zoomcrc.com Join by H.323 162.255.37.11 (US West) 162.255.36.11 (US East) 221.122.88.195 (China) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 209.9.211.110 (Hong Kong) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 645 162 020

1 0

Reminder: Colloquium Speaker Ravi Netravali, tomorrow at 12:30pm
by Emily C. Lawrence 22 Mar '20

22 Mar '20

CS Colloquium Speaker Speaker: Ravi Netravali, University of California, Los Angeles Date: Monday, March 23 - 12:30pm Host: Karthik Narasimhan Location: Remote via Zoom Link: https://princeton.zoom.us/j/645162020 Event page: https://www.cs.princeton.edu/events/25925 ***Due to the developing situation surrounding the COVID-19 virus, this talk will be available for remote viewing. See below for details.*** Title: Improving Stack-Wide Resource Utilization for a Faster Mobile Web Abstract: Mobile web pages are integral to today's society, supporting critical services such as education, e-commerce, and social networking. Despite considerable academic and industrial research efforts, and major improvements over the past decade across the client-side web stack (i.e., networks, device CPUs, and browser engines), page load performance has plateaued and continues to fall short of user performance demands in practice. The consequences of this are far reaching: users abandon pages early, costing content providers billions of dollars in lost revenue; or pages are unusably slow, particularly in developing regions where web pages are often the sole gateway to the aforementioned services. In this talk, I will describe the origin of this performance plateau in the context of serialized page load tasks that preclude effective utilization of the underlying network and CPU resources. Then, I will describe two complementary optimizations that my students and I have developed to eliminate these inefficiencies throughout the page load process and cut mobile load times in half. Key to these optimizations are judicious applications of programming languages (e.g., symbolic execution) and machine learning (e.g., reinforcement learning) techniques that enable us to 1) discover optimization knobs that preserve application correctness, and 2) tune those knobs according to stack-wide signals from the network, device, page, and browser, without developer intervention. I will conclude by describing how these underlying techniques can motivate and address a range of future challenges in networked applications and distributed systems. Bio: Ravi Netravali is an Assistant Professor of Computer Science at UCLA. His research interests are broadly in computer systems and networking, with a recent focus on building practical systems to improve the performance and debugging of large-scale, distributed applications for both end users and developers. His research has been recognized with an NSF CAREER Award, a Google Faculty Research Award, an ACM SoCC Best Paper Award, and an IRTF Applied Networking Research Prize. Prior to joining UCLA, Netravali received a PhD in Computer Science from MIT in 2018. Zoom information: Topic: Ravi Netravali CS Seminar Time: Mar 23, 2020 12:00 PM Eastern Time (US and Canada) Join Zoom Meeting https://princeton.zoom.us/j/645162020 Meeting ID: 645 162 020 One tap mobile +13126266799,,645162020# US (Chicago) +16465588656,,645162020# US (New York) Dial by your location +1 312 626 6799 US (Chicago) +1 646 558 8656 US (New York) +1 253 215 8782 US +1 301 715 8592 US +1 346 248 7799 US (Houston) +1 669 900 6833 US (San Jose) Meeting ID: 645 162 020 Find your local number: https://princeton.zoom.us/u/avcvlf1F3 Join by SIP 645162020(a)zoomcrc.com Join by H.323 162.255.37.11 (US West) 162.255.36.11 (US East) 221.122.88.195 (China) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 209.9.211.110 (Hong Kong) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 645 162 020

1 0

Anders Miltner will present his Pre FPO on Tuesday, March 24, 2020 at 11am
by Nicki Mahler 18 Mar '20

18 Mar '20

Anders Miltner will present his Pre FPO on Tuesday, March 24, 2020 at 11am. To attend the talk, you may log into the following Zoom Meeting: [ https://princeton.zoom.us/j/525520011 | https://princeton.zoom.us/j/525520011 ] The members of his committee are as follows: David Walker (reader, adviser); Kathleen Fisher (reader - Tufts University), Andrew Appel (examiner), Aarti Gupta (examiner), and Zak Kincaid (examiner) All are welcome to attend. Please see abstract below. Bidirectional transformations are groupings of transformations between pairs of data representations. A particularly well-studied class of bidirectional transformations are lenses, bidirectional transformations that satisfy a set of “round-tripping laws”. Lenses can implement serializer/deserializer pairs, parser/pretty printer pairs, and incremental data structure transformers. Domain-specific languages like Boomerang, Augeas, GRoundTram, BiFluX, BiYacc, Brul, BiGUL, and HOBiT allow programmers to write both transformations with a single program, and guarantee the transformations satisfy the round-tripping laws. As long as an expression is well-typed in these languages, the bidirectional transformation that expression represents is a lens. However, writing in domain-specific lens languages can be hard, requiring the user to reason about fiddly details of the transformations within the context of a complex type system. In this work we introduce Optician, which provides an alternative method of developing lenses – synthesis. We develop a synthesis engine for Boomerang, a lens language for string transformations. Pairs of regular expressions serve as types for Boomerang lenses, which transform between the languages of those regular expressions. Instead of manually writing Boomerang lenses, programmers merely need to provide the type of the desired lens and a set of input/output examples describing the lens’s behavior. Optician will then synthesize a lens between the provided types that exhibits the demonstrated behavior. We demonstrate how to synthesize three classes of lenses: bijective lenses, symmetric lenses, and quotient lenses. Bijective lenses encode bijections, symmetric lenses encode bidirectional transformations that permit losing information when going from one format to another, and quotient lenses satisfy the round-tripping laws modulo an equivalence relation. We define a synthesizer that involves two cooperating procedures. One procedure proposes a candidate spaces of lenses, the second procedures searches through that space. To synthesize symmetric lenses, regular expressions are augmented with probability distributions, and the synthesis algorithm aims to avoid information loss. To synthesize quotient lenses, regular expressions are augmented with equivalence information, and the synthesizer generates lenses between the representative elements of the equivalence classes. We evaluate Optician from a variety of benchmarks taken from the lens and synthesis literature, and find that we are able to synthesize all the lenses in our benchmark suite. We have integrated Optician into the Boomerang codebase, enabling users to either synthesize their lenses or write them by hand. Join Zoom Meeting [ https://princeton.zoom.us/j/525520011 | https://princeton.zoom.us/j/525520011 ]

1 0

Colloquium Speaker Ravi Netravali, Monday, Mar 23 at 12:30pm
by Emily C. Lawrence 16 Mar '20

16 Mar '20

CS Colloquium Speaker Speaker: Ravi Netravali, University of California, Los Angeles Date: Monday, March 23 - 12:30pm Host: Karthik Narasimhan Location: Remote via Zoom Link: https://princeton.zoom.us/j/645162020 Event page: https://www.cs.princeton.edu/events/25925 ***Due to the developing situation surrounding the COVID-19 virus, this talk will be available for remote viewing. See below for details.*** Title: Improving Stack-Wide Resource Utilization for a Faster Mobile Web Abstract: Mobile web pages are integral to today's society, supporting critical services such as education, e-commerce, and social networking. Despite considerable academic and industrial research efforts, and major improvements over the past decade across the client-side web stack (i.e., networks, device CPUs, and browser engines), page load performance has plateaued and continues to fall short of user performance demands in practice. The consequences of this are far reaching: users abandon pages early, costing content providers billions of dollars in lost revenue; or pages are unusably slow, particularly in developing regions where web pages are often the sole gateway to the aforementioned services. In this talk, I will describe the origin of this performance plateau in the context of serialized page load tasks that preclude effective utilization of the underlying network and CPU resources. Then, I will describe two complementary optimizations that my students and I have developed to eliminate these inefficiencies throughout the page load process and cut mobile load times in half. Key to these optimizations are judicious applications of programming languages (e.g., symbolic execution) and machine learning (e.g., reinforcement learning) techniques that enable us to 1) discover optimization knobs that preserve application correctness, and 2) tune those knobs according to stack-wide signals from the network, device, page, and browser, without developer intervention. I will conclude by describing how these underlying techniques can motivate and address a range of future challenges in networked applications and distributed systems. Bio: Ravi Netravali is an Assistant Professor of Computer Science at UCLA. His research interests are broadly in computer systems and networking, with a recent focus on building practical systems to improve the performance and debugging of large-scale, distributed applications for both end users and developers. His research has been recognized with an NSF CAREER Award, a Google Faculty Research Award, an ACM SoCC Best Paper Award, and an IRTF Applied Networking Research Prize. Prior to joining UCLA, Netravali received a PhD in Computer Science from MIT in 2018. Zoom information: Topic: Ravi Netravali CS Seminar Time: Mar 23, 2020 12:00 PM Eastern Time (US and Canada) Join Zoom Meeting https://princeton.zoom.us/j/645162020 Meeting ID: 645 162 020 One tap mobile +13126266799,,645162020# US (Chicago) +16465588656,,645162020# US (New York) Dial by your location +1 312 626 6799 US (Chicago) +1 646 558 8656 US (New York) +1 253 215 8782 US +1 301 715 8592 US +1 346 248 7799 US (Houston) +1 669 900 6833 US (San Jose) Meeting ID: 645 162 020 Find your local number: https://princeton.zoom.us/u/avcvlf1F3 Join by SIP 645162020(a)zoomcrc.com Join by H.323 162.255.37.11 (US West) 162.255.36.11 (US East) 221.122.88.195 (China) 115.114.131.7 (India Mumbai) 115.114.115.7 (India Hyderabad) 213.19.144.110 (EMEA) 103.122.166.55 (Australia) 209.9.211.110 (Hong Kong) 64.211.144.160 (Brazil) 69.174.57.160 (Canada) 207.226.132.110 (Japan) Meeting ID: 645 162 020

1 0