eduzhai > Applied Sciences > Computer Science >

Objective Social Choice Using Auxiliary Information to Improve Voting Outcomes

  • Peter
  • (0) Download
  • 20210216
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: How should one combine noisy information from diverse sources to make aninference about an objective ground truth? This frequently recurring, normativequestion lies at the core of statistics, machine learning, policy-making, andeveryday life. It has been called "combining forecasts ", "meta-analysis ", "ensembling ", and the "MLE approach to voting ", among other names. Past studiestypically assume that noisy votes are identically and independently distributed(i.i.d.), but this assumption is often unrealistic. Instead, we assume thatvotes are independent but not necessarily identically distributed and that ourensembling algorithm has access to certain auxiliary information related to theunderlying model governing the noise in each vote. In our present work, we: (1)define our problem and argue that it reflects common and socially relevant realworld scenarios, (2) propose a multi-arm bandit noise model and count-basedauxiliary information set, (3) derive maximum likelihood aggregation rules forranked and cardinal votes under our noise model, (4) propose, alternatively, tolearn an aggregation rule using an order-invariant neural network, and (5)empirically compare our rules to common voting rules and naiveexperience-weighted modifications. We find that our rules successfully useauxiliary information to outperform the naive baselines.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×