Zoom: https://illinois.zoom.us/j/84367822181?pwd=BUZfCKruikpgbLUbsaqrndrN74a4w3.1
Refreshments Provided.
Abstract:
We will discuss two recent developments in scientific processes.
1) Autonomous AI Scientists: Research conducted by autonomous AI scientist systems is rapidly increasing in prevalence. These systems execute the entire research process autonomously, with little or no human intervention. Papers generated by such systems have been accepted at reputable venues, including the ACL conference and ICLR workshops. While papers generated by AI scientist systems appear appealing, we dig deeper into their workflow and investigate whether these systems follow rigorous research methodologies. Using novel experiments designed to mitigate confounding factors, we uncover significant and concerning methodological flaws in the workflow followed by autonomous AI scientists. We also propose a method to detect these problems and provide policy recommendations for journals and conferences: such methodological problems are not detectable from the produced paper alone but can be identified through analysis of the trace logs of the workflow executed by the AI scientists.
2) Lotteries for Funding: Traditional funding decisions (such in the NSF) involve expert reviews followed by panel discussions to determine which proposals are funded. More recently though, citing drawbacks of the traditional approach, a number of funding agencies worldwide have moved towards a different decision model. These agencies have incorporated “partial lotteries” into their decision-making: the review process remains similar to that of the NSF (sometimes omitting panel discussions), but the final decisions introduce a randomized component that still respects reviewers’ evaluations. We will first identify several problems in current implementations of such partial lotteries. We will then present a principled approach to designing improved partial lotteries with strong theoretical guarantees and empirical performance.
The talk will also contain a generous dose of minions.
Bio:
Nihar B. Shah is an Associate Professor in the Machine Learning and Computer Science departments at Carnegie Mellon University (CMU). His research focuses on the Evaluation of Science and the Science of Evaluation. His group develops computational tools with strong theoretical guarantees, and designs and conducts controlled experiments for evidence-based policy design. His work has been used in the review of well over a hundred thousand papers and thousands of proposals, across over 200 venues. He is a recipient of the Young Alumnus Medal from the Indian Institute of Science, a JP Morgan faculty research award, Google Research Scholar Award, an NSF CAREER Award, and the David J. Sakrison memorial prize from EECS Berkeley for a "truly outstanding and innovative PhD thesis." Papers authored by him have won awards from HCOMP, AAMAS, IEEE Data Storage, and several workshops. He loves the minions and making memes.
Part of the Siebel School Speakers Series. Faculty Host: Han Zhao
Meeting ID: 843 6782 2181
Passcode: csillinois
If accommodation is required, please email <erink@illinois.edu> or <communications@cs.illinois.edu>. Someone from our staff will contact you to discuss your specific needs