Computational Intelligence and Neuroscience / 2009 / Article / Tab 3

Research Article

Unsupervised Learning of Overlapping Image Components Using Divisive Input Modulation

Table 3

Details of the training procedure used for each of the benchmarking algorithms tested. In all cases the parameters values listed were those found to produce the best results. Parameter values were kept constant across variations in the task. All algorithms except 𝛽 = 0 . 0 1 use an online learning procedure and hence training time is measured in cycles, whereas for 𝜈 = 0 . 1 training time is measured in epochs. See the caption of Table 2 for further details.

AlgorithmTraining timeIterationsWeight initialisationParameter values

πš‘ 𝚊 πš› πš™ 𝚞 πš› 𝟸 200 000 cyclesn/amean = 1 3 2 , std = 1 8 𝛽 = 0 . 0 0 2 5 , πœ‡ = 0 . 0 5
πœƒ = 0 . 5 20 000 cycles50mean = πš— πš– 𝚏 𝚜 𝚌 , std = 1 2 1 8 , 𝑠 π‘Š = 0 . 5 , 𝑠 π‘Œ =
𝚍 πš’ 2 000 epochsn/amean = 1 3 6 , std = 0 . 0 0 1 𝛽 + = 0 . 2 5 , 𝛽 βˆ’ = 0 . 2 5 none
𝚍 πš’ πš– 20 000 cycles25mean = 1 1 6 , std = 1 6 4 𝛽 = 0 . 0 5 , 𝑠 = 2
𝑠 = 3 20 000 cycles50mean = 𝑠 = 4 , std = 𝚍 πš’ πš– # 1

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.