Crowds, Learning and Incentives
How can machines and people work together to accomplish what neither can alone? Our research focuses on problems at the interplay of learning and incentives in crowd-powered systems. Motivated by applications in community sensing, human computation, citizen science and crowdsourcing, we focus on particular technical challenges such as: How to optimally design incentives for learning / information gathering? How can one piece together a global picture from large numbers of noisy observations of a priori unknown quality?We address these fundamental questions by building on state of the art results in machine learning, probabilistic modeling and game theory. Our research spans various application domains, including community sensing systems for earthquake detection and air-quality monitoring, rebalancing bike sharing systems by steering the behavior of the users, incentivizing users for privacy-tradeoff in information elicitation, teaching and training crowd workers, exploring social aspects of crowdsourcing, and learning optimal pricing policies for online crowdsourcing marketplaces.
Publications
2022 | |
| |
2018 | |
| |
2017 | |
| |
| |
| |
2016 | |
| |
| |
| |
| |
2015 | |
| |
| |
| |
| |
| |
| |
2014 | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
2013 | |
| |
| |
| |
| |
2012 | |
| |
2011 | |
| |
| |
| |
| |
2010 | |
| |
| |
| |
2008 | |
| |
|