Department Events & Announcements

Events

  • Sep
    12

    CS Colloquium - Paul Valiant - "Four Challenges in Learning"

    Department of Computer Science

    4:00 PM 3514, Mudd Hall ( formerly Seeley G. Mudd Library)

    EVENT DETAILS

    Abstract

    In this talk I will motivate and discuss solutions to four learning challenges: How to efficiently optimize beyond the class of convex functions; How to get good performance in deep learning despite overfitting training data; How to efficiently count rare items in data; and, How to make predictions when the future is not like the past.

    The first result shows how to extend polynomial-time optimization algorithms from the classic case of convex optimization to a broader class of "star-convex" functions, which capture some natural non-convex settings in learning. The algorithm is a randomized adaptation of the ellipsoid algorithm which, unexpectedly, uses information from far outside the feasible region. The second result confronts one of the main challenges in deep learning: understanding why "overparameterized" models that can perfectly fit the training data still have such good performance on new data. We introduce a new notion of implicit regularization that arises by comparing stochastic gradient descent to an Ornstein-Uhlenbeck process. The third result considers the problem of rapidly estimating the frequency of a rare and expensive-to-verify occurrence in a database, such as a data error that might require human judgement to detect. We introduce an adaptive algorithm that abandons further investigation of items that look potentially not-rare, without introducing bias in the estimate. Finally, we consider a fundamental and perennial challenge for machine learning: as opposed to assuming that the test set and training set were drawn from the same distribution, suppose we instead know that some different probabilistic process took place to produce the training data and test data. We introduce a new framework based on a semidefinite-programming relaxation to yield almost-optimal linear estimators in the case of mean estimation in the challenging setting where an adversary chooses data values.

    more less

    TIME Thursday, September 12, 2019 at 4:00 PM - 5:00 PM

    LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library)    map it

    ADD TO CALENDAR SHARE

    CONTACT Brianna White    brianna.mello@northwestern.edu EMAIL

    CALENDAR Department of Computer Science

  • Sep
    23

    Welcome & Luncheon for New Full-time Graduate Students

    McCormick School of Engineering and Applied Science

    11:30 AM Ryan Family Auditorium, Technological Institute

    EVENT DETAILS

    TIME Monday, September 23, 2019 at 11:30 AM - 1:30 PM

    LOCATION Ryan Family Auditorium, Technological Institute    map it

    ADD TO CALENDAR SHARE

    CONTACT Northwestern Engineering Events    northwestern-engineering-events@northwestern.edu EMAIL

    CALENDAR McCormick School of Engineering and Applied Science

  • Sep
    27

    CS/PL Guest Speaker - Philip Wadler - (Programming Languages) in Agda = Programming (Languages in Agda)

    Department of Computer Science

    2:00 PM 3514, Mudd Hall ( formerly Seeley G. Mudd Library)

    EVENT DETAILS

    TIME Friday, September 27, 2019 at 2:00 PM - 3:00 PM

    LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library)    map it

    ADD TO CALENDAR SHARE

    CONTACT Brianna White    brianna.mello@northwestern.edu EMAIL

    CALENDAR Department of Computer Science

  • Oct
    7

    CS Colloquium - Ying Wu - "Making Intelligent Seeing Machines"

    Department of Computer Science

    12:00 PM 3514, Mudd Hall ( formerly Seeley G. Mudd Library)

    EVENT DETAILS

    TIME Monday, October 7, 2019 at 12:00 PM - 1:00 PM

    LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library)    map it

    ADD TO CALENDAR SHARE

    CONTACT Brianna White    brianna.mello@northwestern.edu EMAIL

    CALENDAR Department of Computer Science

  • Oct
    15

    The Inaugural Stephen H. Davis Symposium

    McCormick School of Engineering and Applied Science

    10:00 AM Hive Studio, 2-350, Ford Motor Company Engineering Design Center

    EVENT DETAILSmore info

    TIME Tuesday, October 15, 2019 at 10:00 AM - 5:00 PM

    LOCATION Hive Studio, 2-350, Ford Motor Company Engineering Design Center    map it

    ADD TO CALENDAR SHARE

    CONTACT Northwestern Engineering Events    mccormick-events@northwestern.edu EMAIL

    CALENDAR McCormick School of Engineering and Applied Science

  • Oct
    30

    CS+X Colloquium - Caroline Uhler

    Department of Computer Science

    12:00 PM 3514, Mudd Hall ( formerly Seeley G. Mudd Library)

    EVENT DETAILS

    TIME Wednesday, October 30, 2019 at 12:00 PM - 1:00 PM

    LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library)    map it

    ADD TO CALENDAR SHARE

    CONTACT Brianna Mello    brianna.mello@northwestern.edu EMAIL

    CALENDAR Department of Computer Science