Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We define the concept of Bregman continuity and show that a function is an optimal policy function of a discounted dynamic optimization problem if and only if it is Bregman continuous.