After the very successful web seminar on ChatGPT in February we offer you another event on generative ai, this time with a special focus on image generation. And: this will be our first „transatlantic“ web seminar: the second speaker, Marcos Cuzziol of the Media Solution Center, will dial in from Sao Paolo, Brazil. So, do not miss this special web seminar scheduled for July 18th from 2:00 PM to 3:00 PM.
The Web-Seminar will consist of two parts:
Basics (Joseph Pareti)The power of generative AI and collaborative innovations is unleashed through OpenAI’s ChatGPT. It builds upon the foundation laid by transformers and large language models. Pre-training, fine-tuning, reinforcement learning, reward models, and policy optimization contribute to its architecture. Large language models have practical value, with specialized systems like Google med-palm2. Generative AI shows promising performance in causal reasoning, a unique human ability.
Images and generative AI (Marcos Cuzziol)A visual overview on the process of AI image generation: how it works, what’s implied, what to expect in the future.And a guideline for the presentation:
- a deterministic process generating unforeseen results?
- generative adversarial networks: one part of the algorithm competing with (and improving) the other
- Patterns in the data: how data shape the outcome
- What to expect in the near future
Here you can register for free.We look forward to your participation in the web seminar, where we will delve into the exciting world of Generative AI and explore the intricate relationship between images and generative AI.