Singular Learning Theory (2)

Here, we will clarify a road returning from algebraic geometry to statistical learning theory. (1) By the resolution theorem, it is proved that the Gelfand zeta function is meromorphic. (2) The inverse Mellin transform of the zeta function is the state density function. (3) The Laplace transform of the state density function is the partition function. (4) Finally, we elucidate the marginal likelihood using empirical process theory. 

This is the road from algebraic geometry to singular learning theory. The inverse Mellin transform of the zeta function is the state density function, whose Laplace transform is the partition function.

The log likelihood function can be understood as an empirical process defined on algebraic variety, which converges to a Gaussian process in distribution. 

slt202307.pdf

Singular Learning Theory 07

slt202308.pdf

Singular Learning Theory 08

slt202309.pdf

Singular Learning Theory 09

slt202310.pdf

Singular Learning Theory 10

slt202311.pdf

Singular Learning Theory 11

slt202312.pdf

Singular Learning Theory 12

slt202313.pdf

Singular Learning Theory 13

slt202314.pdf

Singular Learning Theory 14

slt202315.pdf

Singular Learning Theory 15