Compressing and Teaching for Low VC-Dimension

@article{Moran2015CompressingAT,
  title={Compressing and Teaching for Low VC-Dimension},
  author={Shay Moran and Amir Shpilka and Avi Wigderson and Amir Yehudayoff},
  journal={2015 IEEE 56th Annual Symposium on Foundations of Computer Science},
  year={2015},
  pages={40-51}
}
In this work we study the quantitative relation between VC-dimension and two other basic parameters related to learning and teaching. Namely, the quality of sample compression schemes and of teaching sets for classes of low VC-dimension. Let C be a binary concept class of size m and VC-dimension d. Prior to this work, the best known upper bounds for both parameters were log(m), while the best lower bounds are linear in d. We present significantly better upper bounds on both as follows. We… CONTINUE READING

Similar Papers

Topics from this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 56 REFERENCES

On the complexity of teaching

VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

Space-bounded learning and the vapnik-chervonenkis dimension

S. Floyd
  • COLT, 1989, pp. 349–364.
  • 1989
VIEW 16 EXCERPTS
HIGHLY INFLUENTIAL

Densité et dimension

VIEW 3 EXCERPTS
HIGHLY INFLUENTIAL

Chervonenkis, “On the uniform convergence of relative frequencies of events to their probabilities.

A. V. Vapnik
  • Theory Probab. Appl.,
  • 1971
VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

A Geometric Approach to Sample Compression

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL