A Distributed Conjugate Gradient Online Learning Method over Networks

Algorithm 1

Distributed Online Conjugate Gradient Algorithm (D-OCG).

(1)	Input: convex set , maximum round number T
(2)	Initialize: , ∀i ∈ V
(3)	for i = 1, …, T do
(4)	The adversary reveals fti, ∀i ∈ V
(5)	Compute the gradients  ∈ ∂fti(xi(t)), ∀i ∈ V
(6)	Compute:
	,
	βi(t) = max{0, βi(t)^PRP},
(7)
	, ∀i ∈ V
(8)	for each i ∈ V do
(9)
(10)
(11)	end for
(12)	end for