$$E(\mathbf{w}) = \frac{1}{|\mathcal{S}|} \sum_{x \in \mathcal{S}} \|f(\mathbf{x}, \mathbf{w}) - h(\mathbf{x})\|^2$$
I was reading about supervised learning in neural networks when i came across this. Can someone explain what the double vertical bar notation means? Is it the frobenius norm squared or simply the L2 norm? Can someone simplify this equation for me.