#27 AI Series: DeepMind - D. Gökay

We are excited to feature Dilara Gökay, who is currently a Research Engineer at Google DeepMind and will discuss "Training Large Scale Video-Native Models", lasting approximately 45 minutes. After the talk, seize the opportunity to connect with fellow AI enthusiasts to share ideas and questions while enjoying free drinks and pizza. Door close by 7.15pm, so please come early! Also, "attend"ing (RSVP) here on Meetup is strictly necessary to be guaranteed entry.
Please note that Meetup has recently been quite keen on promoting its Plus program. However, you are not obligated to purchase it, as both our events and the platform remain free.

Who is this event for?
This event is open to everyone interested in state-of-the-art AI research. We especially design it for students, PhD candidates, academic researchers, and industry professionals with a research focus in machine learning.

Abstract: Real-world understanding necessitates modeling complex temporal and motion cues, yet current image-first approaches often fall short in capturing "what is happening" in favor of "what" is merely present. Furthermore, convincingly demonstrating scaling for pure self-supervised learning from video has remained a challenge, largely because prior evaluation has focused on semantic tasks like action classification. This talk addresses these limitations by introducing a scalable, video-native approach built on masked auto-encoding. We demonstrate that by focusing evaluation on challenging non-semantic 4D vision tasks—such as point and object tracking, camera pose, and depth estimation—MAE with transformer video models actually scales. Specifically, we show consistent performance improvements as the model size is increased from 20 million up to a new industry record of 22 billion parameters, rigorously confirming the benefits of scaling 4D representations.

Bio: Dilara is a Research Engineer at Google DeepMind in London, working on video understanding. She received her MSc in Computer Science from the Technical University of Munich in 2022, specializing in computer vision and graphics as a TEV/DAAD scholar, and her BSc in Computer Engineering from Boğaziçi University, where she graduated with high honors. Previously, she held engineering roles at Facebook Reality Labs, Microsoft, and X (The Everyday Robot Project), as well as a site reliability internship at Google.

We are BLISS e.V., the AI organization in Berlin that connects like-minded individuals who share great interest and passion for the field of machine learning. This winter 2025/26, we will, again, host an exciting speaker series on site in Berlin, featuring excellent researchers from Tübingen AI Center, DeepMind, Microsoft, King's College London, cohere, and ETH Zürich.
Website: https://bliss.berlin
Youtube: https://www.youtube.com/@bliss.ev.berlin

Disclaimer: By attending this event you agree to be photographed.