The Evidence Framework Applied to Classification Networks

  title={The Evidence Framework Applied to Classification Networks},
  author={David J. C. MacKay},
  journal={Neural Computation},
Three Bayesian ideas are presented for supervised adaptive classifiers. First, it is argued that the output of a classifier should be obtained by marginalizing over the posterior distribution of the parameters; a simple approximation to this integral is proposed and demonstrated. This involves a "moderation" of the most probable classifier's outputs, and yields improved performance. Second, it is demonstrated that the Bayesian framework for model comparison described for regression models in… CONTINUE READING
Highly Influential
This paper has highly influenced 59 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 1,158 citations. REVIEW CITATIONS
341 Citations
15 References
Similar Papers


Publications citing this paper.
Showing 1-10 of 341 extracted citations

1,158 Citations

Citations per Year
Semantic Scholar estimates that this publication has 1,158 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 15 references

Probabilistic interpretation of feedforward classification

  • J S.
  • 1989
Highly Influential
1 Excerpt

Bayesian interpolation

  • C.J.
  • Neural Comp. 4(3), 415-447.
  • 1992
1 Excerpt

Active selection of training examples for net

  • 1991

Active selection of training examples for network learning in noiseless environments

  • H. S. Seung, H. Sompolinsky, N. Tishby
  • Dept . Computer Science
  • 1991
1 Excerpt

Similar Papers

Loading similar papers…