Lower Bound Methods and Separation Results for On-Line Learning Models

@article{Maass2004LowerBM,
  title={Lower Bound Methods and Separation Results for On-Line Learning Models},
  author={Wolfgang Maass and Gy{\"o}rgy Tur{\'a}n},
  journal={Machine Learning},
  year={2004},
  volume={9},
  pages={107-145}
}
We consider the complexity of concept learning in various common models for on-line learning, focusing on methods for proving lower bounds to the learning complexity of a concept class. Among others, we consider the model for learning with equivalence and membership queries. For this model we give lower bounds on the number of queries that are needed to learn a concept class $$\mathcal{C}$$ in terms of the Vapnik-Chervonenkis dimension of $$\mathcal{C}$$ , and in terms of the complexity of… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 33 references

On - line learning of geometrical concepts with membership queries

  • W. Bultman, W. Maass
  • Proceedings of the Fourth Annual Workshop on…
  • 1991

On-line learning of geometrical concepts with membership

  • W. Bultman, W. Maass
  • 1991
1 Excerpt

How fast can a threshold gate learn

  • W. Learning. Maass, Turan, Gy
  • 1990

Similar Papers

Loading similar papers…