Goto

Collaborating Authors

 Country



Participatory Personalization in Classification Supplementary Material

Neural Information Processing Systems

The performance of participatory systems will depend on individual reporting decisions. Thus, flat and sequential systems will perform better than a minimal system. The best-case performance of any participatory system will exceed the performance of any of its components. Given a participatory system, we can conduct this evaluation by simulating the parameters in the individual disclosure model shown above. The sequential system outperforms static personalized systems when all group attributes are reported.



Group-FairOnlineAllocationinContinuousTime

Neural Information Processing Systems

However, inmanyapplications including contractual hiringinonlinefreelancing platforms and server allocation in cloud computing systems, the outcome of each action is observed only after a random and action-dependent time.