ARTIFICIAL NEURAL NETWORK MODULE AND SCHEDULING METHOD THEREOF FOR HIGHLY EFFECTIVE PARALLEL PROCESSING

摘要

According to an embodiment of the present invention, an artificial neural network module and scheduling method thereof may enable an artificial neural network specialized for parallel operation of data units to perform high efficient parallel processing by adding a dummy value for performing a dummy operation to an artificial neural network including the irregular characteristics to allow an artificial neural network for performing a regular operation since sizes of weight matrices included in a plurality of convolutional kernels in a plurality of convolutional layers are difference from each other.