Trading off Mistakes and Don't-Know Predictions

@inproceedings{SayediRoshkhar2010TradingOM,
  title={Trading off Mistakes and Don't-Know Predictions},
  author={Amin S. Sayedi-Roshkhar and Morteza Zadimoghaddam and Avrim Blum},
  booktitle={NIPS},
  year={2010}
}
We discuss an online learning framework in which the agent is allowed to say “I don’t know” as well as making incorrect predictions on given examples. We analyze the trade off between saying “I don’t know” and making m istakes. If the number of don’t-know predictions is required to be zero, the model reduces to the well-known mistake-bound model introduced by Littlest one [Lit88]. On the other hand, if no mistakes are allowed, the model reduces to K WIK framework introduced by Li et. al. [LLW08… CONTINUE READING
Highly Cited
This paper has 18 citations. REVIEW CITATIONS