Архитектура многослойного персептрона (MLP): критерии выбора количества скрытых слоев и размера скрытого слоя?

Если у нас есть 10 собственных векторов, то у нас может быть 10 нейронных узлов на входном слое. Если у нас есть 5 выходных классов, то у нас может быть 5 узлов на выходном слое. Но каковы критерии выбора количества скрытых слоев в MLP и сколько нейронных узлы в 1 скрытом слое?

Ответы на вопрос(3)

Решение Вопроса

how many hidden layers?

Error: User Rate Limit Exceededzero скрытые слои разрешатсяlinearly separableError: User Rate Limit Exceeded

Error: User Rate Limit Exceededalways start with one hidden layerError: User Rate Limit Exceeded


How many nodes in the hidden layer?

Error: User Rate Limit ExceededRoTError: User Rate Limit Exceeded

RoTError: User Rate Limit Exceeded

When you begin the model building, err on the side of more nodes in the hidden layer.

Error: User Rate Limit Exceeded

гнилые клубни,Error: User Rate Limit Exceeded

A rule of thumb is for the size of this [hidden] layer to be somewhere between the input layer size ... and the output layer size....

To calculate the number of hidden nodes we use a general rule of: (Number of inputs + outputs) x 2/3

RoTError: User Rate Limit Exceeded

Typically, we specify as many hidden nodes as dimensions [principal components] needed to capture 70-90% of the variance of the input data set.

Error: User Rate Limit ExceededNN FAQError: User Rate Limit Exceeded

Error: User Rate Limit Exceededchoose the number of neurons in the hidden layer based on whether your MLP includes some form of regularization, or early stopping.

The only valid technique for optimizing the number of neurons in the Hidden Layer:

Error: User Rate Limit Exceeded

Error: User Rate Limit Exceeded


Error: User Rate Limit Exceeded

input layerError: User Rate Limit Exceeded

output layerError: User Rate Limit Exceeded

hidden layer: to start, one hidden layerError: User Rate Limit Exceeded

enter image description here

Error: User Rate Limit ExceededError: User Rate Limit Exceeded.

Error: User Rate Limit Exceeded

  1. Chromosome: Vector that defines how many units in each hidden layer (e.g. [20,5,1,0,0] meaning 20 units in first hidden layer, 5 in second, ... , with layers 4 and 5 missing). You can set a limit on the maximum number number of layers to try, and the max number of units in each layer. You should also place restrictions of how the chromosomes are generated. E.g. [10, 0, 3, ... ] should not be generated, because any units after a missing layer (the '3,...') would be irrelevant and would waste evaluation cycles.
  2. Fitness Function: A function that returns the reciprocal of the lowest training error in the cross-validation set of a network defined by a given chromosome. You could also include the number of total units, or computation time if you want to find the "smallest/fastest yet most accurate network".

Error: User Rate Limit Exceeded

  • Pruning: Start with a large network, then reduce the layers and hidden units, while keeping track of cross-validation set performance.
  • Growing: Start with a very small network, then add units and layers, and again keep track of CV set performance.

Error: User Rate Limit Exceeded

Error: User Rate Limit Exceeded

Error: User Rate Limit ExceededError: User Rate Limit ExceededError: User Rate Limit Exceeded

Ваш ответ на вопрос