Noise-injected analog Ising machines enable ultrafast statistical sampling and machine learning

Abstract

Ising machines are a promising non-von-Neumann computational concept for neural network training and combinatorial optimization. However, while various neural networks can be implemented with Ising machines, their inability to perform fast statistical sampling makes them inefficient for training neural networks compared to digital computers. Here, we introduce a universal concept to achieve ultrafast statistical sampling with analog Ising machines by injecting noise. With an opto-electronic Ising machine, we experimentally demonstrate that this can be used for accurate sampling of Boltzmann distributions and for unsupervised training of neural networks, with equal accuracy as software-based training. Through simulations, we find that Ising machines can perform statistical sampling orders-of-magnitudes faster than software-based methods. This enables the use of Ising machines beyond combinatorial optimization and makes them into efficient tools for machine learning and other applications.

References in corpus (5)

• Large-scale photonic Ising machine by spatial light modulation
• Noise-enhanced spatial-photonic Ising machine
• Order-of-magnitude differences in computational performance of analog Ising machines induced by the choice of nonlinearity
• Coherent Ising machines with error correction feedback
• Efficient sampling of ground and low-energy Ising spin configurations with a coherent Ising machine