[!quote] The road to success is paved with mistakes well handled.
- Stanley Marcus
如果你总是成功，就很难知道要改进什么。如果你经常失败，你就不会学到什么是有效的。只有当我们拥有成功和失败的混合体时，我们才能在好的和坏的策略之间形成对比。
In many situations we find that there is a sweet spot in which training is neither too easy nor too hard, and where learning progresses most quickly. For all of these stochastic gradient-descent based learning algorithms, we find that the optimal error rate for training is around 15.87% or, conversely, that the optimal training accuracy is about 85%.
The 85% rule suggests that you should fine-tune the amount of support you use depending on the success rate you’re experiencing.