Cohn, D.: Separating Formal Bounds from Practical Performance in Learning Systems. PhD thesis, University of Washington (1992) Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Distribution-Dependent Vapnik-Chervonenkis Bounds - Vayatis, Azencott (1999) (Correct)
....remarkable fact that this result holds with no assumption on the probability distribution underlying the data. Consequently, VC theory of bounds is considered as a Worst Case theory. This observation is the source of most of the criticisms addressed to VCtheory. It has been argued (see e.g. 4] [5], 9] 17] that VC bounds are loose in general. Indeed, there is an infinite number of situations in which the observed learning curves representing the generalization error of some learning structure are not well described by theoretical VC bounds. 2 In [17] D. Schuurmans criticizes the ....
Cohn, D.: Separating Formal Bounds from Practical Performance in Learning Systems. PhD thesis, University of Washington (1992)
Theory and Practice of Vector Quantizers Trained on Small.. - Tech Report (1992) Self-citation (Cohn) (Correct)
.... extreme cases where this relationship breaks down, but it holds well for all typical cases we have examined, and allows conversion between the non replacement approach and the theoretically examined approach of sampling with replacement (for a more detailed treatment of this relationship, see [3]) 1 It is easy to confuse this issue with that of drawing distinct but identical blocks. To clarify, imagine that each block in an image is labeled by the row and column of the image in which it appears. When drawing without replacement we could still expect to draw many blocks consisting of ....
D. Cohn. Separating formal bounds from practical performance in learning systems. Ph.D. dissertation, Department of Computer Science and Engineering, University of Washington, 1992.
Theory and Practice of Vector Quantizers Trained on Small.. - David Cohn (1992) (8 citations) Self-citation (Cohn) (Correct)
.... extreme cases where this relationship breaks down, but it holds well for all typical cases we have examined, and allows conversion between the non replacement approach and the theoretically examined approach of sampling with replacement (for a more detailed treatment of this relationship, see [3]) 3.2 Relation to worst case bounds For the single image learning experiments, we examined both binary and grayscale images from three sources: photographic images from the USC database, MRI brain scans, and computer generated line drawings. The binary images were generated by halftoning the ....
D. Cohn. Separating formal bounds from practical performance in learning systems. Ph.D. dissertation, Department of Computer Science and Engineering, University of Washington, 1992.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC