OnlineDiscovery - AI and Robotics Badge available

Robotic foundation models

  • * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

  • Date
    14 October 2025
    Timeframe
    18:00 - 19:00 CEST Geneva
    Duration
    60 minutes

      In this talk, I will discuss how large neural network models, similar to those that have enabled AI assistants, coding tools, and chatbots, can enable general robotic skills. Recent progress in the research community has made it possible to train large neural network models for robotic control that enable tasks that were previously impossible, by transferring semantic knowledge from web-scale pretraining and combining it with physical understanding learned from robot data. I will discuss the implications of these technologies on robotic capabilities in the future, provide an overview of recent scientific results, and discuss perspectives on future progress.

      Learning Objectives:

      By the end of the session, participants will be able to:

      • Explain how large neural network models can enable general robotic skills by integrating semantic and physical knowledge.
      • Analyze recent research breakthroughs that have advanced robotic control through large-scale neural models.
      • Evaluate the potential impact of these technologies on the future capabilities and development of robotics.
      Discovery badge available
      Verified badge available for CHF 10
      Learn more
      Share this session with your network
      Discover more from programme stream

      Are you sure you want to remove this speaker?