Dichotomies
={hypothesis h:→{x,o}}
Call:
h(x1,x2,...,xN)=(h(x1),h(x2),...,h(xN))∈{x,o}N
- A dichotomy: hypothesis limited to the eyes of x1,x2,...,xN
- (x1,x2,...,xN): all dichotomies implemented by on the x1,x2,...,xN
||: Candidate for replacing M
Growth Function
(x1,x2,...,xN) depends ont inputs x1,x2,...,xN- growth function: remove dependencies by taking the MAX of all (x1,x2,...,xN)
m(N)=maxx1,x2,...,xN∈(x1,x2,...,xN)Lines in 2D
N m(N) 1 2 2 4 3 max(...,6,8)=8 4 14<2N - Finite, upper bounded by 2N
Growth function for positive rays(1D perceptron)
Growth function for positive intervals
Growth function for convex sets