Lin Fan, Jiacheng Zou: INFORMS preparation seminar

We will have 2 short presentations before INFORMS 2022. Each talk is ~20min.

Lin Fan: The Fragility of Optimized Bandit Algorithms

Abstract: Much of the literature on optimal design of bandit algorithms is based on minimization of expected regret. It is well known that algorithms that are optimal over certain exponential families can achieve expected regret that grows logarithmically in the number of trials, at a rate specified by the Lai-Robbins lower bound. In this paper, we show that when one uses such optimized algorithms, the resulting regret distribution necessarily has a very heavy tail, specifically, that of a truncated Cauchy distribution. Furthermore, for $p>1$, the $p$’th moment of the regret distribution grows much faster than poly-logarithmically, in particular as a power of the total number of trials. We show that optimized UCB algorithms are also fragile in an additional sense, namely when the problem is even slightly mis-specified, the regret can grow much faster than the conventional theory suggests. Our arguments are based on standard change-of-measure ideas, and indicate that the most likely way that regret becomes larger than expected is when the optimal arm returns below-average rewards in the first few arm plays, thereby causing the algorithm to believe that the arm is sub-optimal. To alleviate the fragility issues exposed, we show that UCB algorithms can be modified so as to ensure a desired degree of robustness to mis-specification. In doing so, we also show a sharp trade-off between the amount of UCB exploration and the tail exponent of the resulting regret distribution.

Website: https://linfanf.github.io/

Lin Fan is on the Academic Job Market.

Jiacheng Zou: Using ML to Improve Utilization and Reducing Discards in Deceased Donor Organ Allocation 

Abstract: In the US, over 20% of deceased donor kidneys are discarded after being recovered for transplantation. While cadaver kidneys are allocated based on priority points, many organ offers are being declined and such organs accumulate cold-ischemic times that further reduce their quality. The kidney allocation system allows out-of-sequence offers of kidneys that are in risk of being discarded. Exploiting this lever and working towards improving efficiency and reducing organ discard, we develop a mechanism that (i) predicts whether a kidney is hard-to-place, (ii) recommends how to make out-of-sequence offers for such organs. We estimate the causal effects of this mechanism in a field experiment with organ procurement organizations.