Low Barrier Nanomagnet Design for Binary Stochastic Neurons: Design Challenges for Real Nanomagnets with Fabrication Defects

Abstract

Much attention has been focused on the design of low barrier nanomagnets (LBM), whose magnetizations vary randomly in time owing to thermal noise, for use in binary stochastic neurons (BSN) which are hardware accelerators for machine learning. The performance of BSNs depend on two important parameters: the correlation time associated with the random magnetization dynamics in a LBM, and the spin-polarized pinning current which stabilizes the magnetization of a LBM in a chosen direction within a chosen time. Here, we show that common fabrication defects in LBMs make these two parameters unpredictable since they are strongly sensitive to the defects. That makes the design of BSNs with real LBMs very challenging. Unless the LBMs are fabricated with extremely tight control, the BSNs which use them could be unreliable or suffer from poor yield.

Accepted for publication in IEEE Magnetics Letters

References in corpus (5)

• Spin Transfer Torques
• Intrinsic optimization using stochastic nanomagnets
• Low Barrier Magnet Design for Efficient Hardware Binary Stochastic Neurons
• Magneto-elastic universal logic gate: A non-volatile, error-resilient Boolean logic gate with ultra-low energy-delay product
• Design of a Low Voltage Analog-to-Digital Converter using Voltage Controlled Stochastic Switching of Low Barrier Nanomagnets

Cited by in corpus (12)

• Straintronics: Manipulating the Magnetization of Magnetostrictive Nanomagnets with Strain for Energy-Efficient Applications
• Double Free-Layer Magnetic Tunnel Junctions for Probabilistic Bits
• Quantitative Evaluation of Hardware Binary Stochastic Neurons
• Hardware Design for Autonomous Bayesian Networks
• Probabilistic Circuits for Autonomous Learning: A simulation study
• Robustness and scalability of p-bits implemented with low energy barrier nanomagnets
• Analog Signal Processing Using Stochastic Magnets
• The effect of material defects on resonant spin wave modes in a nanomagnet
• The Strong Sensitivity of the Characteristics of Binary Stochastic Neurons Employing Low Barrier Nanomagnets to Small Geometrical Variations
• Robustness of binary stochastic neurons implemented with low barrier nanomagnets made of dilute magnetic semiconductors
• Sensitivity of the Power Spectra of Magnetization Fluctuations in Low Barrier Nanomagnets to Barrier Height Modulation and Defects
• A Deep Dive into the Computational Fidelity of High Variability Low Energy Barrier Magnet Technology for Accelerating Optimization and Bayesian Problems