Hedge Update Weights

Problem

Perform one multiplicative-weights (Hedge) update and return the new normalized expert weights:

w_new[i] proportional to w[i] * exp(-eta * loss[i])

Hint

Experts that suffer larger losses lose weight exponentially fast.

case 1

in: [[0.5,0.5],[1,0],0.5]

expected: [0.37754066879814546,0.6224593312018546]

case 2

in: [[1,1],[0,0],1]

expected: [0.5,0.5]

case 3

in: [[0.5,0.5],[0,1],2]

expected: [0.8807970779778823,0.11920292202211755]

case 4

in: [[0.25,0.75],[1,1],1]

expected: [0.25,0.75]

Starter codePython

import math
def hedge_update_weights(weights, losses, eta):
    # Your code here
    pass

4 test cases · run your code against them in the editor