I was trying to understand the source code of boundIter() function which computes a bound for the number of iterations as per Proposition 6.6.5 in “Markov Decision Processes”, M. L. Puterman, 1994:
Although the Hajnal measure (k) is defined by Theorem 6.6.6, and its implementation code is clear, I can’t understand the overall meaning behind the formula:
It’s very difficult for me to relate the equation used to calculate maximum iteration to the formula used in the code:
Can you please explain that formula to me, so that I can understand why the logarithm is used and how Hajnal measure (k) can be used to calculate the maximum iteration?
I was trying to understand the source code of boundIter() function which computes a bound for the number of iterations as per Proposition 6.6.5 in “Markov Decision Processes”, M. L. Puterman, 1994:
Pymdptoolbox: https://github.com/sawcordwell/pymdptoolbox/blob/master/src/mdptoolbox/mdp.py
Although the Hajnal measure (k) is defined by Theorem 6.6.6, and its implementation code is clear, I can’t understand the overall meaning behind the formula:
It’s very difficult for me to relate the equation used to calculate maximum iteration to the formula used in the code:
Can you please explain that formula to me, so that I can understand why the logarithm is used and how Hajnal measure (k) can be used to calculate the maximum iteration?
Your Support is much appreciated!