- This event has passed.
Creative AI Webinar
24 Feb 2023 @ 09:00 - 12:00 CET
About the event
The last year has seen an explosion in "creative Al", in particular, the performance of text generation models (such as chatGPT) and image generation models (such as Stable Diffusion, Dalle-2 and Midjourney) has left many outside of the field shocked by the suddenly impressive capabilities of Al.
In this three-hour webinar, we will give an overview of where these models come from, how they work and what can be expected in the future.
- In the first part, we will start by framing what we mean by creative Al. This will be followed by going back to the heart of deep learning: representation learning. We will discuss Auto Encoders and their wider family and show how latent diffusion models work on a high level. We will then show how this can be used to build generative models which are the building blocks of Stable Diffusion and DALL-E-2.
- In the second part, we will discuss language models, some of their history and the motivation for their use in deep learning. We will give an overview of the Transformer architecture and how it has accelerated the progress in language modelling, going from a pre-training task to being able to solve problems without retraining. We will show how this can be taken steps further by augmenting the language model with other general tasks, resulting in something like chatGPT.
- In the end, we will forecast some of the impacts of these technologies and invite to an open discussion about the topic of creative Al.
Prerequisites
The webinar will be on a popular science level, you are not assumed to be a machine learning expert but should know what it is and have a basic knowledge of neural networks.
Registration
You can register for this event at https://events.prace-ri.eu/event/1482/registrations/1097/
Contact
For any questions contact us at training@enccs.se
Follow our Events Schedule
Follow us on Twitter and subscribe to our Newsletter to stay tuned to our events and other news.
————
This training is intended for users established in the European Union or a country associated with Horizon 2020. You can read more about the countries associated with Horizon2020 here https://ec.europa.eu/info/research-and-innovation/statistics/framework-programme-facts-and-figures/horizon-2020-country-profiles_en
Related Events
This project has received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 951732. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and its associated countries .