> Elements of supervised learning
1. Model : how to make prediction
2. Parameters : the things we need to learn from data
3. Objective Function : Training Loss + Regularization
3.1 Training Loss : measures how well model fit on training data.
3.2 Regularization : measures complexity of model.
- Optimizing training loss encourages predictive models.
; Fitting well in training data at least get you close to training data which is hopefully close to the underlying distribution.
- Optimizing regularization encourages simple models.
; Simpler models tends to have smaller variance in future predictions, make prediction stable.
https://www.youtube.com/watch?v=Vly8xGnNiWs
댓글 없음:
댓글 쓰기