Home

Forward Propogation Introduction This is the second in series of 3 deep learning intro posts: Introduction to Deep Learning which introduces the Deep Learning technology background, and presents network’s building blocks and terms. Forward Propogation, which presents the mathematical equations of the prediction path. In this post we wil...

Back Propogation Example In this post we will run a network Training (aka Fitting) example, based on the Back Propogation algorithm explained in the previous post. The example will run a single Back Propogation cycle, to produce 2 outputs: \(\frac{\mathrm{d} C}{\mathrm{d}{b^{[l]}}}\) and \(\frac{\mathrm{d} C}{\mathrm{d}{w^{[l]}}}\) for 1<l&l...

Appendix: Activation Functions Derivation ##Sigmoid Figure 1: Sigmoid Eq. 1a: Sigmoid Function \[\sigma{x}=\frac{1}{1+e^{-x}}\] Eq. 1a: Sigmoid Derivative \[\frac{\partial } {\partial z}\sigma(z)=\frac{\partial } {\partial z}\frac{1}{1+e^{-z}}= -\frac{-e^{-z}}{(1+e^{-z})^2}=-\frac{1-(1+e^{-z})}{(1+e^{-z})^2}=-\sigma(z)^2+\sigma(z)=\sig...

Introduction to Deep Learning This is the first in series of 3 deep learning intro posts: Introduction to Deep Learning which introduces the Deep Learning technology background, and presents network’s building blocks and terms. Introduction to rd Propogation, which presents the mathematical equations of the prediction path. Introduction...

Customization Table of contents Color schemes Custom schemes Define a custom scheme Use a custom scheme Switchable custom scheme Override and completely custom styles Color schemes New Just the Docs supports two color schemes: light (default), and dark. To enable a color scheme, set the color_scheme ...

Batch and Minibatch Introduction Gradient Descent and its variations are the most common algorithms used for fitting the DNN model during the Training phase. The basic formula of Gradient Descent parameter update is presented in Eq. 1: Eq. 1 Gradient Descent \(w_{t+1}=w_t-\alpha \cdot \triangledown L(w)\) Where \(L(w)\) is a Loss function. ...

Back Propogation Introduction This is the third in series of 3 deep learning intro posts: Introduction to Deep Learning which introduces the Deep Learning technology background, and presents network’s building blocks and terms. Forward Propogation, which presents the mathematical equations of the prediction path. Backward Propogation wh...

A Post with Header Image, See Page layout for more examples.