Deep Learning

Deep Learning #

Section renders pages in section as definition list, using title and description. Optional param summary can be used to show or hide page summary

Log Reg Nn
Logistic Regression as a Neural Network # Architecture # 𝐺𝑖𝑣𝑒𝑛 π‘₯ , 𝑦̂ = 𝑃(𝑦 = 1|π‘₯), where 0 ≀ 𝑦̂ ≀ 1 Parameters of logistic regression Input observation,features matrix X Target vector Y Weights w Threshold or bias b Output: 𝑦̂, sigmoid(z) where z = 𝑀 𝑇 *π‘₯ + 𝑏 To get the parameters w and b (i.e. learning), we optimize on: 𝐽(𝑀, 𝑏) = 1/m (βˆ‘ 𝐿(𝑦̂ (𝑖) , 𝑦 (𝑖) ))