This symbol represents the parameters of the model
This symbol represents the parameters of the model. It is a vector containing all the trainable parameters of a neural network. Typically, it includes the weights and biases of a model.
The symbol, \(\theta \in\) \( \htmlClass{sdt-0000000052}{\Theta} \), represents the parameters of the model. It is a vector containing all the trainable parameters of a neural network. Typically, it includes the weights and biases of a model.
Formally, \(\theta\) is an \(n\)-dimensional vector, where \(n\) is the number of parameters - \(\theta \in \mathbb{R^n}\). As an example, consider a simple linear model: \( \htmlClass{sdt-0000000084}{h} \)\((\htmlClass{sdt-0000000103}{u}) = 2u + 3\), we can distinguish to parameters, a slope - 2, and an intercept - 3. We can write this as: \(\theta = (\theta_1, \theta_2) = (2, 3)\).