IEEE Transactions on Neural Networks · 2010 · 200 citations · 25 references
ComputingComputer ScienceArtificial IntelligenceComputer NetworksMachine LearningDeep LearningNeural NetworksPattern RecognitionBackpropagationFeedforward Neural NetworksMultilayer Perceptron
The method introduced in this paper allows for training arbitrarily connected neural networks, therefore, more powerful neural network architectures with connections across layers can be efficiently trained. The proposed method also simplifies neural network training, by using the forward-only computation instead of the traditionally used forward and backward computation.
25