When I first learned RNN, I found lots of articles about it. That’s good news, but most of them focus on the design of the architecture.The network itself is not difficult to understand, the real problem is why it works. What’s the intuition behind it? Why do we need “recurrence”? Unfortunately, few articles explain it clearly. Luckily, the book Deep Learning gives the answers.

Convolutional Neural Networks (CNN) are widely used for image classification, object detection and other tasks related to images or videos in the field of computer vision. Plenty of architectures based on convolution operation have been proposed in recent years, such as AlexNet and ResNet. So, how do CNNs work? Why not use MLP? In this post, we will go through the fundamentals of CNN and several ConvNets to develop a big picture of CNN.

Model Complexity What is model complexity? Perhaps it mainly refers to the number of parameters that the model has. However, it does not mean that a model is complex if it has more parameters, although it is often the case. Generally, a simple model could result in underfitting, while a complex model could lead to overfitting. So, what factors affect the complexity of a model? Well, it is hard to say.

Neural Networks Neural networks are simply function compositions. Feedforward networks $$ y = f(g(x, \theta_g), \theta_f) $$ Recurrent networks $$ y_t = f(y_{t-1}, x_t, \theta) $$ Generally, you can think of neural networks as function approximation machines, where we want to find appropriate $\theta$ to make $f^{*}$ as similar as possible to the true function $f$. In practice, there are lots of things to consider when designing a neural network.

Data visualisation is one of the modules in the first semester. Though I worked with web design and UI for several years, I’d never think about them deeply. As I learned more about design principles and the use of different charts, my aesthetic sense improved greatly. Here I listed some useful resources just for future reference. DLKW Pyramid https://allthingy.com/data-information-knowledge-wisdom/ Chart Junk Data story Below are key points when you structure a data story