Next
Livestream will start soon!
Livestream has already ended.
Presentation has not been recorded yet!
  • title: Multisample Flow Matching: Straightening Flows with Minibatch Couplings
      0:00 / 0:00
      • Report Issue
      • Settings
      • Playlists
      • Bookmarks
      • Subtitles Off
      • Playback rate
      • Quality
      • Settings
      • Debug information
      • Server sl-yoda-v2-stream-006-alpha.b-cdn.net
      • Subtitles size Medium
      • Bookmarks
      • Server
      • sl-yoda-v2-stream-006-alpha.b-cdn.net
      • sl-yoda-v2-stream-006-beta.b-cdn.net
      • 1549480416.rsc.cdn77.org
      • 1102696603.rsc.cdn77.org
      • Subtitles
      • Off
      • English
      • Playback rate
      • Quality
      • Subtitles size
      • Large
      • Medium
      • Small
      • Mode
      • Video Slideshow
      • Audio Slideshow
      • Slideshow
      • Video
      My playlists
        Bookmarks
          00:00:00
            Multisample Flow Matching: Straightening Flows with Minibatch Couplings
            • Settings
            • Sync diff
            • Quality
            • Settings
            • Server
            • Quality
            • Server

            Multisample Flow Matching: Straightening Flows with Minibatch Couplings

            Jul 24, 2023

            Speakers

            AP

            Aram-Alexandre Pooladian

            Speaker · 0 followers

            HB

            Heli Ben-Hamu

            Speaker · 0 followers

            CD

            Carles Domingo-Enrich

            Speaker · 0 followers

            About

            Simulation-free methods for training continuous-time generative models construct probability paths that go between noise distributions and individual data samples. Recent works, such as Flow Matching, derived paths that are optimal for each data sample. However, these algorithms rely on independent data and noise samples, and do not exploit underlying structure in the data distribution for constructing probability paths. We propose Multisample Flow Matching, a more general framework that uses no…

            Organizer

            I2
            I2

            ICML 2023

            Account · 615 followers

            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

            SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient
            05:44

            SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient

            Max Ryabinin, …

            I2
            I2
            ICML 2023 2 years ago

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

            Forward-Backward Gaussian Variational Inference via JKO in the Bures–Wasserstein Space
            05:06

            Forward-Backward Gaussian Variational Inference via JKO in the Bures–Wasserstein Space

            Michael Diao, …

            I2
            I2
            ICML 2023 2 years ago

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

            ESC: Exploration with Soft Commonsense Constraints for Zero-shot Object Navigation
            05:19

            ESC: Exploration with Soft Commonsense Constraints for Zero-shot Object Navigation

            Kaiwen Zhou, …

            I2
            I2
            ICML 2023 2 years ago

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

            Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models
            05:11

            Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models

            Nikhil Kandpal, …

            I2
            I2
            ICML 2023 2 years ago

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

            LeadFL: Client Self-Defense against Model Poisoning in Federated Learning
            05:15

            LeadFL: Client Self-Defense against Model Poisoning in Federated Learning

            Chaoyi Zhu, …

            I2
            I2
            ICML 2023 2 years ago

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

            Incentivizing Exploration with Linear Contexts and Combinatorial Actions
            05:17

            Incentivizing Exploration with Linear Contexts and Combinatorial Actions

            Mark Sellke

            I2
            I2
            ICML 2023 2 years ago

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

            Interested in talks like this? Follow ICML 2023