AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Importance of Objective Functions in PhD Research
I always think about that question actually if what I'm trying to optimize also disclaimer here not because you you see somebody in the after five six years or ten years of finishing their PhD optimizing another objective function this means that if this objective function is the best one or better one to do. Pre-training is important and then fine-tuning is important but because fine- Tuning gives you better results this means that pre- training is not important right um? So everything I was like trying to brainstorm for starts to become obsolete in like six months or somethingUm, it could be that these types of methods that we've been discussing where you kind of want to minimally deviate from your original