On the Relationship between Bayesian Error Bars and the Input Data Density

  title={On the Relationship between Bayesian Error Bars and the Input Data Density},
  author={Christopher K. I. Williams and Cazhaow S. Qazaz and Bishop},
W e investigate the dependence of Bayesian error bars o n the distribution of data in input space. For generalized linear regression models we derive a n upper bound o n the error bars which shows that, in the neighbourhood of the data points, the error bars are substantially reduced f rom their prior values. For regions of high data density we also show that the contribution t o the output variance due to the uncertainty in the weights can exhibit a n approximate inverse proportionality t o… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.


Publications citing this paper.
Showing 1-4 of 4 extracted citations

Density-Dependent Quantized Least Squares Support Vector Machine for Large Data Sets

IEEE Transactions on Neural Networks and Learning Systems • 2017
View 5 Excerpts
Highly Influenced

Density-dependent quantized kernel least mean square

2016 International Joint Conference on Neural Networks (IJCNN) • 2016
View 1 Excerpt

Continuous and Discrete Time Survival Analysis: Neural Network Approaches

2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society • 2007
View 1 Excerpt


Publications referenced by this paper.
Showing 1-3 of 3 references

Bayesian Interpolation

J C.PvlacKayD.
Neural Computation, • 1992
View 3 Excerpts
Highly Influenced

Novelty Detection and Neural Network

ValidationChris, M.., BishopNeural, ScienceAston, UniversityBirmingham

Bayesian Interpolation ” , Neural Computation , 4 ( 3 ) , 415 - 447

C. PvlacKayD.J.

Similar Papers

Loading similar papers…