The Interplay Between Stability and Regret in Online Learning

@article{Saha2012TheIB,
  title={The Interplay Between Stability and Regret in Online Learning},
  author={Ankan Saha and Prateek Jain and Ambuj Tewari},
  journal={CoRR},
  year={2012},
  volume={abs/1211.6158}
}
This paper considers the stability of online learning algorithms and its implications for learnability (bounded regret). We introduce a novel quantity called forward regret that intuitively measures how good an online learning algorithm is if it is allowed a one-step look-ahead into the future. We show that given stability, bounded forward regret is equivalent to bounded regret. We also show that the existence of an algorithm with bounded regret implies the existence of a stable algorithm with… CONTINUE READING
6 Citations
31 References
Similar Papers

References

Publications referenced by this paper.
Showing 1-10 of 31 references

http://www-stat.wharton.upenn.edu / rakhlin/papers/online learning.pdf

  • Alexander Rakhlin
  • Lecture notes on online learning,
  • 2009
Highly Influential
4 Excerpts

Similar Papers

Loading similar papers…