Ativação ELU
ELU, ou Unidade Linear Exponencial, é uma função de ativação used in artificial redes neurais to introduce non-linearity into the model. It is particularly valued for its ability to mitigate the ‘dying ReLU’ problem, which occurs when neurons output zero para todas as entradas, tornando-se efetivamente inativas e deixando de aprender.
A função ELU é definida matematicamente da seguinte forma:
For an input x, the ELU activation function is:
ELU(x) = x, if x > 0 ELU(x) = α * (e^x - 1), if x ≤ 0
Here, α is a hyperparameter that determines the value of the output for negative inputs. The exponential component for negative inputs allows ELU to produce outputs that are non-zero and smooth, which helps in maintaining a mean output close to zero. This property is an advantage over the standard ReLU função, que produz zero para todas as entradas negativas.
Usando ELU em aprendizado profundo models has been shown to accelerate learning and improve accuracy in certain tasks, especially when dealing with deep architectures. It retains all the benefits of ReLU while providing a gradient for negative inputs, which can lead to better convergence during training.
Em resumo, ELU funções de ativação provide a robust alternative to traditional activation functions, particularly in deep neural networks, by addressing some of their inherent limitations.