12 years ago @LukeBornn was a stats grad student, today he is a part owner of multiple pro sports franchises around the world. Had the chance to hear about that journey last week. One of the sharpest guys in the business, and one of the nicest:
https://t.co/qBtTwAc9ZU
I just put out a preprint with @lucaswu123 , @martypute , and Tim Chan of our paper "Learning Risk Preferences in Markov Decision Processes: an Application to the Fourth Down Decision in Football"!
https://t.co/BhHJgPfEnF
Please check it out and share!
My sister wrote a fantastic op-ed about Evan McMullin and the Utah senate race that just got published in the Salt Lake Tribune! Check it out: https://t.co/TOh4cLfyA9
(Late) update - we won! I'm grateful for the competition committee, the other finalists and my coauthors, Tim Chan, Albert Loa and @nsandholtz. For those interested in the case, it's forthcoming at INFORMS Transactions on Education! #informs2022
For anyone living in Utah, the voter registration deadline is coming up on October 28th. Register to vote! Here's the Utah voter registration website: https://t.co/LknCZ0dOlJ
@brit_covey2 I'm a fellow timpview alum helping some students @Choco_Trail@natehawk2 on a punt return decision analysis at the Y. long shot but if you'd be willing to share what aspects of a punt return situation inform your decision-making we'd be super appreciative
We model observed human decisions from an experiment as sample paths from a Bayesian optimization procedure where the governing acquisition function is unknown. We propose a probabilistic framework to learn the subjects' latent acquisition functions.
Some joint work with @LukeBornn, @yohskay, and Maurice A. Smith is now available in advance publications in Bayesian Analysis #BAjournal!
link: https://t.co/gfz2bbNeNS
I presented on our inverse optimization analysis of the fourth down decision in one of friday's #NESSIS21 sessions. If you missed it it's up on youtube now, hope you'll take a look! https://t.co/6IFB5IDdBH (joint work with @lucaswu123 , @martypute , and Tim Chan)
@sertalpbilal@StatsbyLopez Thanks Sertalp, and thanks for the nice summary of the talk. Three research isn't actually based on the arxiv paper you mentioned early in the , but I hope to have a paper up on arxiv soon for it :) we're getting there
I like @nsandholtz’s link between the variability in commute time and what coaches are maximizing as they think about fourth down choices #NESSIS21
Depending on risk profile, the same decision can have three unique recommendations
Just looking at WR times by gender for each stroke, seems like it'd be generally optimal to have men on Breast and Fly, women on Back and Free (I wonder if US had done this if they could've medaled!). Anyhow it was interesting to see diff teams w/ diff strategies
Just watched the 4x100 mixed swim medley. Fun to watch, and to think about from an analytics perspective---knapsack problem where you select athletes and their strokes, minimizing total time, w/ constraint of 2 men and 2 women and 1 stroke per athlete.