#  Into the Unknown: Assigning Reviewers to Papers with Uncertain Affinities 

 



####  calendar\_today Date and Time 

 **February 3, 2023** 

 12:00PM - 01:30PM EST 

####  pin\_drop Location 

 **SEC 1.413 https://goo.gl/maps/UjUiWMGZCEGh5qMr9**  



 

 



 

 **Friday, 1-2pm in SEC 1.413** with [Justin Payan](https://urldefense.proofpoint.com/v2/url?u=https-3A__justinpayan.github.io_&d=DwMFaQ&c=WO-RGvefibhHBZq3fL85hQ&r=ZOP6tLIqLOHbdgCvrXjUlPta0tw7K_-ivqiItQhh6LQ&m=2O9WSSIB3JcCTwKMqg9ZgJaktYweixBvwl6cTbfmTu7IfJu7eoT4hJnIuXeTpDbX&s=GyoGWbiSgJ4IxdPIzYGdj29Kj7hBrrjXXEaZGLx9XQQ&e=) coming to speak in-person all the way from UMass Amherst on:

 *Into the Unknown: Assigning Reviewers to Papers with Uncertain Affinities*

Peer review cannot work unless qualified and interested reviewers are assigned to every paper. Nearly all automated reviewer assignment approaches estimate real-valued affinity scores for each paper-reviewer pair that serve as proxies for the predicted quality of a future review; conferences then assign reviewers to maximize the utilitarian welfare of the assignment. This procedure does not account for noise in affinity score computation --- reviewers can only bid on a small number of papers, and textual similarity models are inherently probabilistic estimators. In this talk, we will explore the case when paper-reviewer affinity scores are estimated using a probabilistic model. Using these probabilistic estimates, we bound the scores with high probability and maximize the worst-case utilitarian welfare for a reviewer allocation. We will discuss multiple ways to estimate probabilistic affinity scores, and how to robustly maximize welfare in these models. Our general approach can be used to integrate a large variety of probabilistic reviewer-paper affinity models into reviewer assignment, opening the door to a much more robust peer review process.

 

 



 

 See also:- [ Seminar history ](/taxonomycalendarseminar/seminar-history)
 
 

 Share on:- [     Facebook ](#)
- [     Twitter ](#)
- [     Linkedin ](#)
 


 Save: [ Add to calendar calendar\_today ](https://econcs.seas.harvard.edu/node/1664208/event-feed.ics)  Copy link link