A Nonvolatile All-Spin Nonbinary Matrix Multiplier: An Efficient Hardware Accelerator for Machine Learning

We propose and analyze a compact and non-volatile nanomagnetic (all-spin) nonbinarymatrix multiplier performing the multiply-and-accumulate (MAC) operation using two magnetic tunnel junctions (MTJs) - one activated by strain to act as the multiplier and the other activated by spin-orbit torque pulses to act as a domain wall (DW) synapse that performs the operation of the accumulator. Each MAC operation can be performed in similar to 5 ns and the energy dissipated per operation is similar to 500 aJ. This provides a very useful hardware accelerator for machine learning and artificial intelligence tasks that often involve the multiplication of large matrices. The nonvolatility allows the matrix multiplier to be embedded in powerful non-von- Neumann architectures. It also allows all computing to be done at the edge while reducing the need to access the cloud, thereby making artificial intelligence more resilient against cyberattacks.

Automated summary

This paper proposes a compact and nonvolatile nanomagnetic nonbinary matrix multiplier that serves as a useful hardware accelerator for machine learning and artificial intelligence tasks. It can be embedded in non-von-Neumann architectures and reduces reliance on the cloud, making AI more resilient against cyberattacks.