Kaiming 初期化
カイミング初期化、別名 He 初期化, is a technique used to initialize the weights of ニューラルネットワーク, particularly those employing 活性化関数 like ReLU (Rectified Linear Unit). Proper 重みの初期化 is crucial as it can significantly affect the convergence speed and overall performance of the network during training.
The method was introduced by Kaiming He and his collaborators in a 2015 paper. It aims to address the issues of vanishing and 爆発勾配 that can occur during the training of deep networks. These issues arise when weights are initialized too small or too large, leading to ineffective learning.
カイミング初期化では、重みは ガウス分布 with a mean of 0 and a variance that is inversely proportional to the number of input units (fan-in) for each neuron. Specifically, the weights are initialized using the formula:
w ~ N(0, sqrt(2 / fan_in))Where w represents the weight, N denotes a 正規分布, and fan_in is the number of input connections to the neuron. This approach helps maintain a balanced distribution of activations throughout the layers of the network, promoting effective gradient flow during backpropagation.
Kaiming Initialization is particularly effective in deep networks with ReLU and its variants but can be adapted for other activation functions with slight modifications. As a result, it has become a standard practice in training 深層学習 モデルにおいて、より高速で安定した収束率に寄与します。