Corpus ID: 12246009

Consistency versus Realizable H-Consistency for Multiclass Classification

@inproceedings{Long2013ConsistencyVR,
  title={Consistency versus Realizable H-Consistency for Multiclass Classification},
  author={Philip M. Long and R. Servedio},
  booktitle={ICML},
  year={2013}
}
A consistent loss function for multiclass classification is one such that for any source of labeled examples, any tuple of scoring functions that minimizes the expected loss will have classification accuracy close to that of the Bayes optimal classifier. While consistency has been proposed as a desirable property for multiclass loss functions, we give experimental and theoretical results exhibiting a sequence of linearly separable data sources with the following property: a multiclass… Expand
14 Citations
Bayes Consistency vs. H-Consistency: The Interplay between Surrogate Loss Functions and the Scoring Function Class
  • 1
  • Highly Influenced
  • PDF
On the Consistency of AUC Pairwise Optimization
  • 50
  • PDF
On the Consistency of Top-k Surrogate Losses
  • 7
  • PDF
Multiclass Classification Calibration Functions
  • 14
  • Highly Influenced
  • PDF
Learning with Rejection
  • 88
  • PDF
Multi-Class Deep Boosting
  • 49
  • PDF
Quantifying Learning Guarantees for Convex but Inconsistent Surrogates
  • 3
  • PDF
Calibrated Surrogate Losses for Adversarially Robust Classification
  • PDF
Consistent Estimators for Learning to Defer to an Expert
  • 16
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 21 REFERENCES
Random classification noise defeats all convex potential boosters
  • 86
  • PDF
Decision Theoretic Generalizations of the PAC Model for Neural Net and Other Learning Applications
  • D. Haussler
  • Computer Science, Mathematics
  • Inf. Comput.
  • 1992
  • 941
  • PDF
Fisher Consistency of Multicategory Support Vector Machines
  • Yufeng Liu
  • Computer Science, Mathematics
  • AISTATS
  • 2007
  • 84
  • PDF
On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines
  • 2,119
  • PDF
Towards Optimal One Pass Large Scale Learning with Averaged Stochastic Gradient Descent
  • W. Xu
  • Computer Science
  • ArXiv
  • 2011
  • 143
  • Highly Influential
  • PDF
On the Consistency of Multiclass Classification Methods
  • 266
  • Highly Influential
  • PDF
Towards good practice in large-scale learning for image classification
  • 186
  • PDF
Statistical Analysis of Some Multi-Category Large Margin Classification Methods
  • Tong Zhang
  • Mathematics, Computer Science
  • J. Mach. Learn. Res.
  • 2004
  • 242
  • Highly Influential
  • PDF
SOME INFINITY THEORY FOR PREDICTOR ENSEMBLES
  • 159
  • PDF
...
1
2
3
...