2017년 11월 28일 화요일

[ML] Elements of supervised learning

> 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