liubi32326的博客

Recipe of Deep Learning、Good Results on Training Data、Good Results on Testing Data、why Dropout work？

Notation、Backpropagation、Forward pass、Backward pass、Summary

Three Steps for Deep Learning、Fully Connect Feedforward Network、Matrix Operation、Output Layer as Multi-Class Classifier、Example Application

Step 1: Function Set、Step 2: Goodness of a Function、Step 3: Find the best function、Why not Logistic Regression + Square Error、Discriminative v.s. Generative、Multi-class Classification(3 Class)、Limitation of Logistic Regression、Deep Learning!

Classification 概念、Example Application、How to do Classification、Why not Regesssion、Probability from Class - Feature、Probability from Class、How’s the results?、Modifying Model、Three Steps、Probability Distribution

Different Optimizers、Optimizers Application、Adam vs SGDM、Towards Improving Adam、Towards Improving SGDM、RAdam vs SWATS、k step forward, 1 step back、Can we look into the future?、Do you really know your optimizer?、Something helps optimization…、learned &Advices

Tuning your learning rates、Stochastic Gradient Descent、Feature Scaling、Gradient Descent：Theory、More Limitation of Gradient Descent

宝可梦预测第一次尝试、Model 结果探讨、宝可梦预测第二次尝试、误差来源探究、处理偏差/方差的注意事项