AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Foundation Model Triad: Fine Tuning Prompt Engineering and Reinforcement Learning With Human Feedback
How do I provide feedback or training that is specifically for me, not the whole general model? That will be something interesting. And also observability, right? Explainability. This is the thing we have kind of taken for granted in the old like supervised learning. But that's also the biggest weakness of deep learning model like that.