Neural Network Primitives

The chapters in this module, in reading order.

#	Chapter
00	Neural networks in kid words — the cat-spotting robot
01	The XOR problem - why one straight rule cannot learn feature interaction
02	Activation functions — the bend that saves the rule pile
03	The forward pass — one prediction, end to end
04	Weight initialization — where the rule pile starts
05	Loss functions — measuring wrongness
06	Backpropagation
07	SGD, batches, epochs — how the nudge actually happens
08	Vanishing gradients — when the nudge dies in deep piles
09	Optimizers — smart nudging that actually moves
10	Regularization — keeping the rule pile honest
11	Scaling laws — bigger pile, more photos, in the right ratio
12	Batch norm vs layer norm — same cleanup, different axis
13	Honest admission — what we still do not know