Talk on Data Engineering and AI

Artificial Intelligence has become ubiquitous in today’s technological landscape due to its strong generalization capabilities. This enables applying similar methods and models to a wide range of problems. However, treating AI models as off-the-shelf, black-box solutions severely limits their potential. In our talk “A recipe for training neural networks” we will explore a systematic approach to training an AI model proposed by Andrej Karpathy, the Co-founder of OpenAI, that we aspire to follow in our research to maximize the performance of AI models. We will contrast this idealistic view with the reality of delivery projects, which often have to cope with financial and time constraints.

Beteiligung: Speaker

Thema: Data Engineering and AI

Wann: 14:15 Uhr

Manjusha Pattadkal

Data Scientist, XITASO

Julian Gleißner

Robotics Researcher, XITASO

Remote via Zoom
link after registration