Massively parallel probabilistic computing with sparse Ising machines
نویسندگان
چکیده
Inspired by the developments in quantum computing, building domain-specific classical hardware to solve computationally hard problems has received increasing attention. Here, introducing systematic sparsification techniques, we demonstrate a massively parallel architecture: sparse Ising Machine (sIM). Exploiting sparsity, sIM achieves ideal parallelism: its key figure of merit - flips per second scales linearly with number probabilistic bits (p-bit) system. This makes up 6 orders magnitude faster than CPU implementing standard Gibbs sampling. Compared optimized implementations TPUs and GPUs, delivers 5-18x speedup In benchmark such as integer factorization, can reliably factor semiprimes 32-bits, far larger previous attempts from D-Wave other solvers. Strikingly, beats competition-winning SAT solvers (by 4-700x runtime reach 95% accuracy) solving 3SAT problems. Even when sampling is made inexact using clocks, find correct ground state further speedup. The problem encoding techniques introduce be applied Machines (classical quantum) architecture present used for scaling demonstrated 5,000-10,000 p-bits 1,000,000 or more through analog CMOS nanodevices.
منابع مشابه
Segmenting Text Images With Massively Parallel Machines
Image segmentation, the partitioning of an image into meaningful parts, is a major concern of any computer vision system. The meaningful parts of a text image are lines of text, words and characters. In this paper, the segmentation of pages of text into lines of text and lines of text into characters on a parallel machine will be examined. Using a parallel machine for text image segmentation al...
متن کاملMassively Parallel Computing in Economics
This paper discusses issues related to parallel computing in Economics. It highlights new methodologies and resources that are available for solving and estimating economic models and emphasizes situations when they are useful and others where they are impractical. Two examples illustrate the different ways parallel methods can be employed to speed computation as well as their limitations.
متن کاملObject-Oriented Programming for Massively Parallel Machines
Large, robust massively parallel programs that are understandable (and therefore maintain-able) are not yet a reality. Such programs require a programming methodology that minimizes the conceptual diierences between the program and the domain addressed by the program, encourages reusability, and still produces robust programs that are readily maintained and reasoned about. This paper proposes t...
متن کاملMassively Parallel Mixed-Signal VLSI Kernel Machines
Recently it has been shown that a simple learning paradigm, the support vector machine (SVM), outperforms some of the most elaborately tuned expert systems and neural networks in object recognition tasks. In run-time, the SVM operates by computing a kernelbased distance between the object’s vector at the input and a set of support vectors selected from the training set, and weighting the result...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Nature electronics
سال: 2022
ISSN: ['2520-1131']
DOI: https://doi.org/10.1038/s41928-022-00774-2