Differences in numbers of remaining hidden units.
收藏数据链接：
官方服务：
资源简介：
Each table entry contains the mean “μ” number of hidden neurons nh after each pruning iteration, and their standard deviation (σ) in parenthesis. The results were averaged over 10 simulations. The model initially had nv = 13, nh = 70, and nw = nv × nh = 910 weights. During each of the three pruning iterations half of the weights were removed according to different criteria. nw describes the remaining weights at the given phase of pruning.
创建时间：
2021-10-11
