METHOD FOR CALCULATING AN OUTPUT OF A NEURONAL NETWORK

• Main Authors: VOGEL SEBASTIAN, SCHORN CHRISTOPH
• Format: Patent
• Language: Chinese; English
• Published: 05.07.2019
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention relates to a method for calculating an output of a neuronal network (100), comprising the steps of generating a first neuronal network (100-1) having discrete edge weights (Wb) from a neuronal network (100) having precise edge weights (Wprec) by stochastic rounding; generating a second neuronal network (100-2) having discrete edge weights (Wb) from the neuronal network (100) having precise edge weights (Wprec) by stochastic rounding; and calculating an output by adding the output of the first neuronal network (100-1) and of the second neuronal network (100-2). 本发明涉及一种用于计算神经网络（100）的输出的方法，所述方法具有以下步骤：通过随机取整，由具有精确的棱边权重（W）的神经网络（100）产生具有离散的棱边权重（W）的第一神经网络（100-1）；通过随机取整，由具有精确的棱边权重（W）的神经网络（100）产生具有离散的棱边权重（W）的第二神经网络（100-2）；以及通过将所述第一神经网络（100-1）的输出和所述第二神经网络（100-2）的输出相加来计算输出。