A very very short introduction
What (x,y) values will minimize my function ?
Convexity guarantees a single global minimum
Follow the slope !
Start from some point x
$$ x_{n+1} = x_{n} - [ \nabla^2 f(x_n) ]^{-1} \nabla f (x_n) $$
Computing and inverting the Hessian can be very costly, quasi-Newton work around it
Example loss function for regression:
$$ L = \parallel y - f_{w}(x) \parallel^2 $$
$$ L = \sum\limits_{i=1}^{N} (y_i - f_w(x_i))^2$$
Deconvolution
Inpainting
Denoising
$$ y = \bm{A} x + n $$
A is non-invertible or ill-conditioned
$$ L = \parallel y - \bm{A} x \parallel^2 + R(x) $$
$$ R(x) = \lambda \parallel x \parallel_2 $$
Checkout the numerical tours
$$ L = \parallel y - \bm{A} x \parallel^2 + R(x) $$
$$ R(x) = \lambda \parallel \Phi x \parallel_1 $$