Structured models of physics, objects, and scenes
a talk by
Peter Battaglia
DeepMind
Thursday, 18 February 2021
11:00 – 12:00 CET
FREE Online Zoom Webinar
For connection information please contact: [email protected]
This talk will describe various ways of using structured machine learning models for predicting complex physical dynamics, generating realistic objects, and constructing physical scenes. The key insight is that many systems can be represented as graphs with nodes connected by edges, which can be processed by graph neural networks and transformer-based models. By considering the underlying structure of the problem, and imposing inductive biases within our models that reflect them, we can often achieve more accurate, efficient, and generalizable performance than if we avoid using principled assumptions.
Peter Battaglia is a research scientist at DeepMind working on approaches for reasoning about and interacting with complex systems.
Trustworthy Complex and Intelligent Systems Webinar Series
This series is a collaboration between the European Safety, Reliability & Data Association (ESReDA), the ETH Zürich Chair of Intelligent Maintenance Systems, ETH Risk Center, the Norwegian Research Center for AI Innovation (NorwAI) and DNV GL.
Webinars will be run monthly throughout 2021 exploring the themes of trust, ethics and applications of AI and novel technology in complex and safety critical intelligent systems.
Next webinar:
with
Asun St. Clair, DNV GL
19 March 2021, 16:00 CET
For further information contact:
[email protected]
[email protected]
Elizabeth Traiger
[email protected]
Sponsored by
