Robustness of Binary Stochastic Neurons Implemented With Low Barrier Nanomagnets Made of Dilute Magnetic Semiconductors

Abstract

Binary stochastic neurons (BSNs) are excellent hardware accelerators for machine learning. A popular platform for implementing them is low- or zero-energy barrier nanomagnets possessing in-plane magnetic anisotropy (e.g., circular disks or quasi-elliptical disks with very small eccentricity). Unfortunately, small geometric variations in the lateral shapes of such nanomagnets can produce large changes in the BSN response times if the nanomagnets are made of common metallic ferromagnets (Co, Ni, Fe) with large saturation magnetization. In addition, the response times become very sensitive to initial conditions, i.e., the initial magnetization orientation. In this letter, we show that if the nanomagnets are made of dilute magnetic semiconductors with much smaller saturation magnetization than common metallic ferromagnets, then the variability in their response times (due to shape variations and variation in the initial condition) is drastically suppressed. This significantly reduces the device-to-device variation, which is a serious challenge for large-scale neuromorphic systems. A simple material choice can, therefore, alleviate one of the most aggravating problems in probabilistic computing with nanomagnets.