- talks - lists.cs.princeton.edu

Prof. Vikram Adve, U Illinois. March 21. 4pm. CS Small Auditorium
by Margaret Martonosi 20 Mar '18

20 Mar '18

Wednesday March 21, 4pm CS Small Auditorium TITLE: HPVM: Simplifying Heterogeneous Parallel Programming Prof. Vikram Adve, University of Illinois, Urbana-Champaign ABSTRACT: Heterogeneous parallel systems are attractive because they use specialized computing elements, including GPUs, vector hardware, FPGAs, and domain-specific accelerators, that can greatly improve energy efficiency, performance, or both, compared with traditional homogeneous systems. A major drawback, however, is that programming and reasoning about heterogeneous parallel systems is highly challenging for application developers, language and compiler designers, and hardware engineers. In the HPVM project, we are developing a unified parallel program representation, Heterogeneous Parallel Virtual Machine, that aims to capture the wide range of parallelism available in such systems. We use the HPVM representation for three different purposes: (1) a compiler intermediate representation (IR), (2) a virtual ISA for shipping (virtual) object code, and (3) a representation for runtime scheduling. We have implemented the HPVM IR and associated compiler tools as an extension of the LLVM compiler infrastructure. In this talk, I will describe the HPVM representation, experimental results demonstrating the ability of HPVM to achieve performance comparable to separately hand-tuned code for GPUs and vector hardware, and to enable highly flexible scheduling policies. I will conclude with a brief discussion of how well-chosen system interfaces can enable highly programmable, specialized hardware, and the influence of such interfaces on key layers of the system stack. BIO: Vikram Adve is the Donald B. Gillies Professor and Interim Head of Computer Science at the University of Illinois at Urbana-Champaign, and has affiliate appointments with the Coordinated Sciences Laboratory and the Information Trust Institute. Adve's research interests lie in developing and using compiler techniques to improve the performance, programmability and reliability of computer systems. Adve and his Ph.D. student, Chris Lattner, co-designed the LLVM Compiler Infrastructure, which enabled novel compilation capabilities for a wide range of languages. LLVM is widely used in industry today, ranging from mobile devices (e.g., in iOS and Android) to supercomputers (e.g., at Cray and NVIDIA) to data centers (e.g., at Google). Adve, Lattner and Evan Cheng received the ACM Software System Award in 2012 for co-developing LLVM (making UIUC the only University to win this Award twice: the Mosaic Web Browser won it in 1995). Adve has won a ten-year-retrospective Most Influential Paper award at CGO 2004, and distinguished paper awards at several conferences including PLDI 2005, SOSP 2007 and ICSE 2011. Adve's Ph.D. student, Robert Bocchino, won the 2010 ACM SIGPLAN Outstanding Dissertation Award and another Ph.D. student, John Criswell, won Honorable Mentions for both the 2014 ACM SIGOPS Dissertation Award and the 2014 ACM Doctoral Dissertation Award. Adve is a Fellow of the ACM and was named a University Scholar at the University of Illinois in 2015. Margaret Martonosi H. T. Adams '35 Professor of Computer Science. Director, Keller Center for Innovation in Engineering Education Princeton University ■ mrm(a)princeton.edu ■ http://www.princeton.edu/~mrm

1 1

Fisher Yu FPO - March 27, 2018 at 10:30 am Room CS 402
by Barbara A. Mooring 20 Mar '18

20 Mar '18

Good afternoon, Fisher Yu will present his FPO on, March 27, 2018 at 10:30 am in Room CS 402. The members of his Committee are: Thomas Funkhouser - Princeton University Adam Funkelstein - Princeton University Olga Russakovsky - Princeton University All are welcomed to attend. Title: Pixel-Level Prediction: From Geometry to Semantics Abstract: Pixel-level prediction generalizes a wide range of computer vision tasks including semantic image segmentation and dense depth prediction. They are fundamental for image recognition, receiving continual attention from the community. However, although they share common traits that may admit a general solution, they are usually studied in isolation because of different domain characteristics. This thesis aims to study the essential problems behind those tasks and shed light on a general framework. This thesis starts with an algorithm that can predict plausible depth from almost identical images based on geometric optimization. The motion between those images is called ``Accidental Motion''. The analysis of accidental motion shows that motion optimization has special convexity properties. It leads to a reconstruction pipeline that can produce a plausible dense depth map for the reference image, which is shown to enable depth based camera effects. The second part then studies learning pixel representation to predict semantic properties based on the single reference image. Previous works usually use learned upsampling to recover the pixel-level information. This work proposes to use Dilated Convolution to transform the classification networks such that high-resolution prediction is achieved without upsampling. Dilated Convolution can also render an exponential increase in receptive field, which is ideal for learning global context. A context module is proposed based on this property that can improve the network performance significantly and consistently. Dilation is still a standard component in the state-of-the-art method for semantic image segmentation. The further study of dilated residual networks shows that same high-resolution prediction can also improve image classification results. This indicates no essential network architecture difference exists between image classification and segmentation. Further inspection of class activation maps and layer responses uncover peculiar gridding patterns and their cause. This finding leads to new designs of convolutional networks that can remove the gridding artifacts and produce activations with better spatial consistency. The new networks can improve the performance of both image classification and semantic segmentation. The presented method and results may inspire new research in building a unified framework for image recognition of geometry and semantics. Barbara Mooring Interim Graduate Coordinator Computer Science Department Princeton University

1 0

Colloquium Speaker-Cathy Wu, Monday, March 26
by Emily C. Lawrence 20 Mar '18

20 Mar '18

***PLEASE NOTE THE DIFFERENT TIME AND LOCATION CS and EE Department Colloquium Speaker Cathy Wu, from the University of California at Berkeley Monday, March 26, 2018 - 4:30pm Bowen Hall - Room 222 Host: Prof. Prateek Mittal (EE) & Szymon Rusinkiewicz (CS) “Mixed-autonomy mobility: scalable learning and optimization” Abstract: How will self-driving cars change urban mobility? This talk describes contributions in machine learning and optimization critical for enabling mixed-autonomy mobility, the gradual and complex integration of automated vehicles into the existing transportation system. The talk first explores and quantifies the potential impact of a small fraction of automated vehicles on low-level traffic flow dynamics such as traffic congestion, using novel techniques in model-free deep reinforcement learning. Second, the talk presents generic reinforcement learning techniques for improved variance reduction, developed for large-scale control systems such as traffic networks and robotic manipulation. To anchor this work in a broader mobility planning context, the coordination of automated vehicles relies on accurate traffic flow sensing. To this end, a new convex optimization method for cellular network measurements from AT&T for all of California is introduced to address a flow estimation problem previously believed to be intractable. Finally, automated vehicles are expected to increase transportation demand through a phenomenon called induced demand. To address this, joint work with Microsoft Research is presented, which provides theoretical justification for the application of widely used clustering algorithms to ridesharing problems, designed to mitigate the strain on existing infrastructure. Together, these contributions demonstrate, through principled learning and optimization methods, that a small number of vehicles and sensors can be harnessed for significant impact on urban mobility. Bio: Cathy Wu is a PhD candidate in machine learning in Electrical Engineering and Computer Sciences (EECS) at UC Berkeley, the Berkeley Artificial Intelligence Research lab, Berkeley DeepDrive , California PATH, and the Berkeley RISELab . She is interested in developing principled computational tools to enable reliable and complex decision-making for critical societal infrastructure, such as transportation systems. Cathy received her Masters and Bachelors degrees in EECS from MIT. She is the recipient of several fellowships including the NSF graduate fellowship, the Berkeley Chancellor's fellowship, the NDSEG fellowship, and the Dwight David Eisenhower graduate fellowship. Her work was acknowledged by several awards, including the 2016 IEEE ITSC Best Paper Award and the 2017 ITS Outstanding Graduate Student Award. Her leadership, in particular as the Research Lead of the Learning Traffic Team at Berkeley, was recognized by numerous awards and invitations, such as the 2017 IEEE Leaders Summit and multiple NSF early-career investigator workshops on cyber-physical systems. Throughout her career, Cathy has collaborated or interned broadly across fields, including civil engineering, mechanical engineering, urban planning, and public policy, and institutions, including OpenAI , Microsoft Research, the Google Self-Driving Car Team, AT&T, Facebook, and Dropbox. As the founder and Chair of the Interdisciplinary Research Initiative within the ACM Future of Computing Academy, she is actively working to create international programs to further enable and support interdisciplinary research in computing.

1 0

Colloquium Speaker-Danqi Chen, Monday, March 26
by Emily C. Lawrence 20 Mar '18

20 Mar '18

CS Department Colloquium Speaker Danqi Chen from Stanford University Monday, March 26, 2018 - 12:30pm Computer Science - Room 105 Host: Barbara Engelhardt “Knowledge from Language via Deep Understanding” Abstract: Almost all of human knowledge is now available online, but the vast majority of it is principally encoded in the form of human language explanations. In this talk, I explore novel neural network approaches that open up opportunities for getting a deep understanding of natural language text. First, I show how distributed representations enabled the building of a smaller, faster and more accurate dependency parser for finding the structure of sentences. Then I show how related neural technologies can be used to improve the construction of knowledge bases from text. However, maybe we don't need this intermediate step and can directly gain knowledge and answer people's questions from large textbases ? In the third part, I explore this possibility by directly reading text with a simple yet highly effective neural architecture for question answering. Bio: Danqi Chen is a PhD student in Computer Science at Stanford University, working with Christopher Manning on deep learning approaches to natural language processing. Her research centers on how computers can achieve a deep understanding of human language and the information it contains. Danqi received Outstanding Paper Awards at ACL 2016 and EMNLP 2017, a Facebook Fellowship, a Microsoft Research Women’s Fellowship and an Outstanding Course Assistant Award from Stanford. Previously, she received her B.E. in Computer Science from Tsinghua University.

1 0

Prof. Vikram Adve, U Illinois. March 21. 4pm. CS Small Auditorium.
by Margaret Martonosi 17 Mar '18

17 Mar '18

Prof. Vikram Adve from UIUC will be “in the neighborhood” next week and giving a seminar about his recent work. Please join us! Wednesday March 21, 4pm CS Small Auditorium TITLE: HPVM: Simplifying Heterogeneous Parallel Programming Prof. Vikram Adve, University of Illinois, Urbana-Champaign ABSTRACT: Heterogeneous parallel systems are attractive because they use specialized computing elements, including GPUs, vector hardware, FPGAs, and domain-specific accelerators, that can greatly improve energy efficiency, performance, or both, compared with traditional homogeneous systems. A major drawback, however, is that programming and reasoning about heterogeneous parallel systems is highly challenging for application developers, language and compiler designers, and hardware engineers. In the HPVM project, we are developing a unified parallel program representation, Heterogeneous Parallel Virtual Machine, that aims to capture the wide range of parallelism available in such systems. We use the HPVM representation for three different purposes: (1) a compiler intermediate representation (IR), (2) a virtual ISA for shipping (virtual) object code, and (3) a representation for runtime scheduling. We have implemented the HPVM IR and associated compiler tools as an extension of the LLVM compiler infrastructure. In this talk, I will describe the HPVM representation, experimental results demonstrating the ability of HPVM to achieve performance comparable to separately hand-tuned code for GPUs and vector hardware, and to enable highly flexible scheduling policies. I will conclude with a brief discussion of how well-chosen system interfaces can enable highly programmable, specialized hardware, and the influence of such interfaces on key layers of the system stack. BIO: Vikram Adve is the Donald B. Gillies Professor and Interim Head of Computer Science at the University of Illinois at Urbana-Champaign, and has affiliate appointments with the Coordinated Sciences Laboratory and the Information Trust Institute. Adve's research interests lie in developing and using compiler techniques to improve the performance, programmability and reliability of computer systems. Adve and his Ph.D. student, Chris Lattner, co-designed the LLVM Compiler Infrastructure, which enabled novel compilation capabilities for a wide range of languages. LLVM is widely used in industry today, ranging from mobile devices (e.g., in iOS and Android) to supercomputers (e.g., at Cray and NVIDIA) to data centers (e.g., at Google). Adve, Lattner and Evan Cheng received the ACM Software System Award in 2012 for co-developing LLVM (making UIUC the only University to win this Award twice: the Mosaic Web Browser won it in 1995). Adve has won a ten-year-retrospective Most Influential Paper award at CGO 2004, and distinguished paper awards at several conferences including PLDI 2005, SOSP 2007 and ICSE 2011. Adve's Ph.D. student, Robert Bocchino, won the 2010 ACM SIGPLAN Outstanding Dissertation Award and another Ph.D. student, John Criswell, won Honorable Mentions for both the 2014 ACM SIGOPS Dissertation Award and the 2014 ACM Doctoral Dissertation Award. Adve is a Fellow of the ACM and was named a University Scholar at the University of Illinois in 2015. Margaret Martonosi H. T. Adams '35 Professor of Computer Science. Director, Keller Center for Innovation in Engineering Education Princeton University ■ mrm(a)princeton.edu ■ http://www.princeton.edu/~mrm

1 0

Colloquium Speaker-He He, Happening TODAY at 12:30
by Emily Lawrence 13 Mar '18

13 Mar '18

CS Department Colloquium Speaker He He from Stanford University Tuesday, March 13, 2018 - 12:30pm Computer Science - Room 105 Host: Barbara Engelhardt “Learning Interactive Agents” Abstract: AI has made huge advancement into our daily life and increasingly we require intelligent agents that work intimately with people in a changing environment. However, current systems mostly work in a passive mode: waiting for requests from users and processing them one at a time. An interactive agent must handle real-time, sequential inputs and actively collaborate with people through communication. In this talk, I will present my recent work addressing challenges in real-time language processing and collaborative dialogue. The first part involves making predictions with incremental inputs. I will focus on the application of simultaneous machine interpretation and show how we can produce both accurate and prompt translations. Then, I will present my work on building agents that collaborate with people through goal-oriented conversation. I will conclude by discussing future directions towards adaptive, active agents. Bio: He He is a postdoctoral researcher at Stanford University. She earned her Ph.D. in Computer Science at the University of Maryland, College Park. She is interested in natural language processing and machine learning. Her research focuses on building intelligent agents that work in a changing environment and interact with people, with an emphasis on language-related problems. Specific applications include dependency parsing, simultaneous machine interpretation, and goal-oriented dialogue. She is the recipient of the 2016 Larry S. Davis doctoral dissertation award.

1 0

CORRECTION-Colloquium Speaker-He He still scheduled for Tuesday, 3/16
by Emily Lawrence 12 Mar '18

12 Mar '18

Correction, He He’s talk is still scheduled for TOMORROW, Tuesday, March 13, 2018 at 12:30pm barring a potential weather emergency. CS Department Colloquium Speaker He He from Stanford University Tuesday, March 13, 2018 - 12:30pm Computer Science - Room 105 Host: Barbara Engelhardt “Learning Interactive Agents” Abstract: AI has made huge advancement into our daily life and increasingly we require intelligent agents that work intimately with people in a changing environment. However, current systems mostly work in a passive mode: waiting for requests from users and processing them one at a time. An interactive agent must handle real-time, sequential inputs and actively collaborate with people through communication. In this talk, I will present my recent work addressing challenges in real-time language processing and collaborative dialogue. The first part involves making predictions with incremental inputs. I will focus on the application of simultaneous machine interpretation and show how we can produce both accurate and prompt translations. Then, I will present my work on building agents that collaborate with people through goal-oriented conversation. I will conclude by discussing future directions towards adaptive, active agents. Bio: He He is a postdoctoral researcher at Stanford University. She earned her Ph.D. in Computer Science at the University of Maryland, College Park. She is interested in natural language processing and machine learning. Her research focuses on building intelligent agents that work in a changing environment and interact with people, with an emphasis on language-related problems. Specific applications include dependency parsing, simultaneous machine interpretation, and goal-oriented dialogue. She is the recipient of the 2016 Larry S. Davis doctoral dissertation award.

1 0

Colloquium Speaker-He He, TODAY at 12:30
by Emily C. Lawrence 12 Mar '18

12 Mar '18

CS Department Colloquium Speaker He He from Stanford University Tuesday, March 13, 2018 - 12:30pm Computer Science - Room 105 Host: Barbara Engelhardt “Learning Interactive Agents” Abstract: AI has made huge advancement into our daily life and increasingly we require intelligent agents that work intimately with people in a changing environment. However, current systems mostly work in a passive mode: waiting for requests from users and processing them one at a time. An interactive agent must handle real-time, sequential inputs and actively collaborate with people through communication. In this talk, I will present my recent work addressing challenges in real-time language processing and collaborative dialogue. The first part involves making predictions with incremental inputs. I will focus on the application of simultaneous machine interpretation and show how we can produce both accurate and prompt translations. Then, I will present my work on building agents that collaborate with people through goal-oriented conversation. I will conclude by discussing future directions towards adaptive, active agents. Bio: He He is a postdoctoral researcher at Stanford University. She earned her Ph.D. in Computer Science at the University of Maryland, College Park. She is interested in natural language processing and machine learning. Her research focuses on building intelligent agents that work in a changing environment and interact with people, with an emphasis on language-related problems. Specific applications include dependency parsing, simultaneous machine interpretation, and goal-oriented dialogue. She is the recipient of the 2016 Larry S. Davis doctoral dissertation award.

1 0

Reminder: Colloquium Speaker-He He, tomorrow at 12:30
by Emily C. Lawrence 12 Mar '18

12 Mar '18

CS Department Colloquium Speaker He He from Stanford University Tuesday, March 13, 2018 - 12:30pm Computer Science - Room 105 Host: Barbara Engelhardt “Learning Interactive Agents” Abstract: AI has made huge advancement into our daily life and increasingly we require intelligent agents that work intimately with people in a changing environment. However, current systems mostly work in a passive mode: waiting for requests from users and processing them one at a time. An interactive agent must handle real-time, sequential inputs and actively collaborate with people through communication. In this talk, I will present my recent work addressing challenges in real-time language processing and collaborative dialogue. The first part involves making predictions with incremental inputs. I will focus on the application of simultaneous machine interpretation and show how we can produce both accurate and prompt translations. Then, I will present my work on building agents that collaborate with people through goal-oriented conversation. I will conclude by discussing future directions towards adaptive, active agents. Bio: He He is a postdoctoral researcher at Stanford University. She earned her Ph.D. in Computer Science at the University of Maryland, College Park. She is interested in natural language processing and machine learning. Her research focuses on building intelligent agents that work in a changing environment and interact with people, with an emphasis on language-related problems. Specific applications include dependency parsing, simultaneous machine interpretation, and goal-oriented dialogue. She is the recipient of the 2016 Larry S. Davis doctoral dissertation award.

1 0

EE Faculty candidate Alexandros Daglis, Monday March 12
by Scott Karlin 09 Mar '18

09 Mar '18

EE SEMINAR SERIES Speaker: Alexandros Daglis (EPFL) Title: Network-Centric Computing for Online Services Day: Monday, March 12 , 2018 Time: 4:30 pm Room: B205 Engineering Quadrangle Host: Prof. David Wentzlaff Abstract: Modern datacenters offer an abundance of online services to billions of daily users. The most demanding online services come with tight response latency requirements, forcing service providers to keep their massive datasets memory-resident, distributed across the datacenter’s servers. Every user request accesses the memory of hundreds of servers; therefore, fast access to the aggregate memory pool is of crucial importance for service quality. The entity binding all the memory resources together is the datacenter network. Unfortunately, despite the dramatic evolution of datacenter fabrics over the past decade, networking still incurs a latency overhead of tens of microseconds to every remote memory access, making memory pooling impractical. In this talk, I will propose a holistic network-centric system redesign to enable memory pooling in future datacenters, which includes (i) a specialized lightweight network stack; (ii) on-chip integration of the network interface logic; and (iii) new network operations with richer, end-to-end semantics. I will highlight the role and demonstrate the effect of each of these three key features with systems built throughout my dissertation. Bio: Alexandros (Alex) Daglis is a sixth-year PhD student at EPFL, advised by Prof. Babak Falsafi and Prof. Edouard Bugnion. His research interests lie in rack-scale computing and datacenter architectures. Alex advocates tighter integration and co-design of network and compute resources as a necessary approach to tackling the performance overheads associated with inter-node communication in scale-out architectures. He has been a founding member of Scale-Out NUMA, an architecture, programming model, and communication protocol for low-latency, distributed in-memory processing. Scale-Out NUMA has been prototyped and awarded a US patent. As an intern at HP Labs, Alex worked on the design of The Machine’s unique memory subsystem. More details can be found on his CV .

1 0