Dandurand, F., Berthiaume, V., & Shultz, T. R. (in press). A systematic comparison of flat and standard Cascade-Correlation using a student-teacher network approximation task. Connection Science.
Cascade-correlation (cascor) networks grow by recruiting hidden units to adjust their computational power to the task being learned. The standard cascor algorithm recruits each hidden unit on a new layer, creating deep networks. In contrast, the flat cascor variant adds all recruited hidden units on a single hidden layer. Student-teacher network approximation tasks were used to investigate the ability of flat and standard cascor networks to learn the input-output mapping of other, randomly initialized flat and standard cascor networks. For low-complexity approximation tasks, there was no significant performance difference between flat and standard student networks. Contrary to the common belief that standard cascor does not generalize well due to cascading weights creating deep networks, we found that both standard and flat cascor generalized well on problems of varying complexity. On high-complexity tasks, flat cascor networks had fewer connection weights and learned with less computational cost than standard networks did.
Abstracts, papers, chapters, and other documents are posted on this site as an efficient way to distribute reprints. The respective authors and publishers of these works retain all of the copyrights to this material. Anyone copying, downloading, bookmarking, or printing any of these materials agrees to comply with all of the copyright terms. Other than having an electronic or printed copy for fair personal use, none of these works may be reposted, reprinted, or redistributed without the explicit permission of the relevant copyright holders.
Author Posting. (c) Taylor &
Francis, 2007.
This is the author's version of the work. It is posted here
by permission of Taylor & Francis for personal use, not for redistribution.
The definitive version was published in Connection Science,
Volume 19 Issue 3, September 2007.
doi:10.1080/09540090701528951
(http://dx.doi.org/10.1080/09540090701528951)
To obtain a PDF author's version of this particular article, signal your agreement with these copyright terms by clicking on the statement below.
I agree with all of these copyright terms PDF 496KB