Unidade Linear Gated (GLU)
Uma Unidade Linear Gated (GLU) é uma função de ativação de rede neural designed to enhance the model’s ability to capture complex relationships in data by incorporating gating mechanisms. Introduced in a paper by Yann N. Dauphin et al. in 2017, GLUs help in improving the performance of deep learning architectures, particularly in tasks related to processamento de linguagem natural (PLN) e outros dados sequenciais.
As GLUs operam combinando transformações lineares com gates que controlam o fluxo de informações. A fórmula básica para uma GLU pode ser expressa como:
GLU(x) = (W_1 * x) ⊗ σ(W_2 * x)In this formula, W_1 and W_2 are weight matrices, x is the input data, and σ represents the sigmoid activation function. The output of the GLU is the element-wise product of a transformação linear and a gating mechanism, which allows the model to learn which features to focus on while ignoring others.
The gating mechanism provides a way to control the information from the input that gets passed forward through the network, allowing for more effective learning and improved gradient flow. This is particularly useful in deep networks where gradientes que desaparecem pode ser um problema significativo.
GLUs are often used in combination with other neural network layers, such as convolutional or recurrent layers, to enhance their performance. They can be seen as an evolution of traditional funções de ativação like ReLU, as they add a layer of complexity and adaptability that helps in various applications, including language modeling, machine translation, and more.