NeuronFlow: A Hybrid Neuromorphic – Dataflow Processor Architecture for AI Workloads
Autor: | Gokturk Cinserin, Amirreza Yousefzadeh, Orlando Miguel Pires Dos Reis Moreira, R.-J. Zwartenkot, P. Qiao, Jonathan Tapson, A. Kapoor, Fabian Chersi, M. Lindwer, Mina A. Khoei |
---|---|
Rok vydání: | 2020 |
Předmět: |
0209 industrial biotechnology
020901 industrial engineering & automation Neuromorphic engineering Artificial neural network Computer architecture Dataflow Computer science 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing 02 engineering and technology Architecture Dataflow architecture Microarchitecture |
Zdroj: | AICAS |
Popis: | We present a novel computing architecture which combines the event-based and compute-in-network principles of neuromorphic computing with a traditional dataflow architecture. The result is a fine-grained dynamic dataflow system which avoids the coding issues intrinsic to spiking systems, and is suitable for both procedural workload and deep neural network (DNN) inference. The architecture is particularly suitable for computation of sparse CNNs and low-latency applications. We present results from GrAIOne, the first chip designed using the NeuronFlow architecture, which has 200 704 neurons implemented in a 28nm HPC + process. |
Databáze: | OpenAIRE |
Externí odkaz: |