# Weighted Classification Cascades for Optimizing Discovery Significance in the HiggsML Challenge

@inproceedings{Mackey2014WeightedCC, title={Weighted Classification Cascades for Optimizing Discovery Significance in the HiggsML Challenge}, author={L. Mackey and J. Bryan and Man Yue Mo}, booktitle={HEPML@NIPS}, year={2014} }

We introduce a minorization-maximization approach to optimizing common measures of discovery significance in high energy physics. The approach alternates between solving a weighted binary classification problem and updating class weights in a simple, closed-form manner. Moreover, an argument based on convex duality shows that an improvement in weighted classification error on any round yields a commensurate improvement in discovery significance. We complement our derivation with experimental… CONTINUE READING

#### Topics from this paper.

7 Citations

#### References

##### Publications referenced by this paper.

SHOWING 1-8 OF 8 REFERENCES

Overlapped Discrete Multitone Modulation for High Speed Copper Wire Communication

- Computer Science
- 1995

288

Eilam Gross, and Ofer Vitells Asymptotic formulae for likelihoodbased tests of new physics

- 2011

Estimating Divergence Functionals and the Likelihood Ratio by Convex Risk Minimization

- Mathematics, Computer Science
- 2010

360

On Divergences and Informations in Statistics and Information Theory

- Computer Science, Mathematics
- 2006

417