Hardware-aware $insitu$ Boltzmann machine learning using stochastic magnetic tunnel junctions

Abstract

One of the big challenges of current electronics is the design and implementation of hardware neural networks that perform fast and energy-efficient machine learning. Spintronics is a promising catalyst for this field with the capabilities of nanosecond operation and compatibility with existing microelectronics. Considering large-scale, viable neuromorphic systems however, variability of device properties is a serious concern. In this paper, we show an autonomously operating circuit that performs hardware-aware machine learning utilizing probabilistic neurons built with stochastic magnetic tunnel junctions. We show that $insitu$ learning of weights and biases in a Boltzmann machine can counter device-to-device variations and learn the probability distribution of meaningful operations such as a full adder. This scalable autonomously operating learning circuit using spintronics-based neurons could be especially of interest for standalone artificial-intelligence devices capable of fast and efficient learning at the edge.

References in corpus (16)

• Deep Learning in Neural Networks: An Overview
• Stochastic p-bits for Invertible Logic
• Thermal fluctuations of magnetic nanoparticles
• Intrinsic optimization using stochastic nanomagnets
• Large-angle, gigahertz-rate random telegraph switching induced by spin-momentum transfer
• Implementing Bayesian Networks with Embedded Stochastic MRAM
• Weighted p-bits for FPGA implementation of probabilistic circuits
• Low Barrier Magnet Design for Efficient Hardware Binary Stochastic Neurons
• Autonomous Probabilistic Coprocessing with Petaflips per Second
• An atomic Boltzmann machine capable of on-chip learning
• Subnanosecond Fluctuations in Low-Barrier Nanomagnets
• Experimental Demonstration of Probabilistic Spin Logic by Magnetic Tunnel Junctions
• Training a Probabilistic Graphical Model with Resistive Switching Electronic Synapses
• Hardware Design for Autonomous Bayesian Networks
• Probabilistic Circuits for Autonomous Learning: A simulation study
• Voltage-driven Building Block for Hardware Belief Networks

Cited by in corpus (25)

• A full-stack view of probabilistic computing with p-bits: devices, architectures and algorithms
• Probabilistic computing with p-bits
• Roadmap for Unconventional Computing with Nanotechnology
• CMOS + stochastic nanomagnets: heterogeneous computers for probabilistic inference and learning
• Training Deep Boltzmann Networks with Sparse Ising Machines
• Experimental demonstration of an integrated on-chip p-bit core utilizing stochastic Magnetic Tunnel Junctions and 2D-MoS2 FETs
• Efficient Probabilistic Computing with Stochastic Perovskite Nickelates
• Accelerated Quantum Monte Carlo with Probabilistic Computers
• Probabilistic-Bits based on Ferroelectric Field-Effect Transistors for Stochastic Computing
• Enhanced Convergence in p-bit Based Simulated Annealing with Partial Deactivation for Large-Scale Combinatorial Optimization Problems
• The Strong Sensitivity of the Characteristics of Binary Stochastic Neurons Employing Low Barrier Nanomagnets to Small Geometrical Variations
• Double-Free-Layer Stochastic Magnetic Tunnel Junctions with Synthetic Antiferromagnets
• Physics-inspired Ising Computing with Ring Oscillator Activated p-bits
• Robustness of binary stochastic neurons implemented with low barrier nanomagnets made of dilute magnetic semiconductors
• Heisenberg machines with programmable spin-circuits
• Superior probabilistic computing using operationally stable probabilistic-bit constructed by manganite nanowire
• GPU-accelerated simulated annealing based on p-bits with real-world device-variability modeling
• Machine Learning Quantum Systems with Magnetic p-bits
• Superparamagnetic and Stochastic-Write Magnetic Tunnel Junctions for High-Speed True Random Number Generation in Advanced Computing
• Noise-based Local Learning using Stochastic Magnetic Tunnel Junctions
• Metrics for spin-based computing
• Ground-State Probabilistic Logic with the Simplest Binary Energy Landscape for Probabilistic Computing
• Self-Adaptive Ising Machines for Constrained Optimization
• Energy-Efficient p-Bit-Based Fully-Connected Quantum-Inspired Simulated Annealer with Dual BRAM Architecture
• Fe- and Co-based magnetic tunnel junctions with AlN and ZnO spacers