Next
Spotlight talks
Livestream will start soon!
Livestream has already ended.
Presentation has not been recorded yet!
  • title: Economical use of second-order information in training machine learning models
      0:00 / 0:00
      • Report Issue
      • Settings
      • Playlists
      • Bookmarks
      • Subtitles Off
      • Playback rate
      • Quality
      • Settings
      • Debug information
      • Server sl-yoda-v3-stream-007-alpha.b-cdn.net
      • Subtitles size Medium
      • Bookmarks
      • Server
      • sl-yoda-v3-stream-007-alpha.b-cdn.net
      • sl-yoda-v3-stream-007-beta.b-cdn.net
      • 1211500432.rsc.cdn77.org
      • 1991924839.rsc.cdn77.org
      • Subtitles
      • Off
      • English (auto-generated)
      • Playback rate
      • Quality
      • Subtitles size
      • Large
      • Medium
      • Small
      • Mode
      • Video Slideshow
      • Audio Slideshow
      • Slideshow
      • Video
      My playlists
        Bookmarks
          00:00:00
            Economical use of second-order information in training machine learning models
            • Settings
            • Sync diff
            • Quality
            • Settings
            • Server
            • Quality
            • Server

            Economical use of second-order information in training machine learning models

            Dec 13, 2019

            Speakers

            DG

            Donald Goldfarb

            Speaker · 0 followers

            About

            Stochastic gradient descent (SGD) and variants such as Adagrad and Adam, are extensively used today to train modern machine learning models. In this talk we will discuss ways to economically use second-order information to modify both the step size (learning rate) used in SGD and the direction taken by SGD. Our methods adaptively control the batch sizes used to compute gradient and Hessian approximations and and ensure that the steps that are taken decrease the loss function with high probabilit…

            Organizer

            N2
            N2

            NIPS 2019

            Account · 966 followers

            Categories

            Mathematics

            Category · 2.4k presentations

            AI & Data Science

            Category · 10.8k presentations

            About NIPS 2019

            Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.

            Like the format? Trust SlidesLive to capture your next event!

            Professional recording and live streaming, delivered globally.

            Sharing

            Recommended Videos

            Presentations on similar topic, category or speaker

            Invited Talk: Equivariant Multilayer Perceptrons
            40:48

            Invited Talk: Equivariant Multilayer Perceptrons

            Siamak Ravanbakhsh

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Training a Variational Network for use on 3D High Resolution MRI Data in 1 Day
            19:02

            Training a Variational Network for use on 3D High Resolution MRI Data in 1 Day

            Christian Kames

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Poster Spotlights 2
            21:54

            Poster Spotlights 2

            Guneet Dhillon, …

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            A sleep-wake detection algorithm for memory-constrained wearable devices: Change Point Decoder
            09:54

            A sleep-wake detection algorithm for memory-constrained wearable devices: Change Point Decoder

            Ayse Cakmak

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Contributed Talk: Understanding Thermodynamic Variational Inference
            13:38

            Contributed Talk: Understanding Thermodynamic Variational Inference

            Rob Brekelmans

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Causality and Exoplanets
            35:11

            Causality and Exoplanets

            Bernhard Schölkopf

            N2
            N2
            NIPS 2019 5 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Interested in talks like this? Follow NIPS 2019