公开(公告)号：US11429862B2

公开(公告)日：2022-08-30

申请号：US16133446

申请日：2018-09-17

Applicant: SRI International

Inventor： Sek Meng Chai ,  Aswin Nadamuni Raghavan ,  Samyak Parajuli

IPC: G06N3/08 ,  G06N3/10 ,  G06N3/04

Abstract: Techniques are disclosed for training a deep neural network (DNN) for reduced computational resource requirements. A computing system includes a memory for storing a set of weights of the DNN. The DNN includes a plurality of layers. For each layer of the plurality of layers, the set of weights includes weights of the layer and a set of bit precision values includes a bit precision value of the layer. The weights of the layer are represented in the memory using values having bit precisions equal to the bit precision value of the layer. The weights of the layer are associated with inputs to neurons of the layer. Additionally, the computing system includes processing circuitry for executing a machine learning system configured to train the DNN. Training the DNN comprises optimizing the set of weights and the set of bit precision values.