blueblisters on Every Model Learned by Gradient Descent Is Approximately a Kernel Machine · Hackerly

Comment on Every Model Learned by Gradient Descent Is Approximately a Kernel Machine parent

blueblisters 5y

y(x) = \sum_{(x_i, y_i) \in L} a_i K(x, x_i) + b.

HN feature request: parse inline latex math please