Stochastic rounding and reduced-precision fixed-point arithmetic for solving neural ordinary differential equations

DOI10.1098/rsta.2019.0052zbMath1462.65081arXiv1904.11263OpenAlexW3098574217WikidataQ92754330 ScholiaQ92754330MaRDI QIDQ4993504

Michael Hopkins, Steve B. Furber, Mantas Mikaitis, Dave R. Lester

Publication date: 15 June 2021

Published in: Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1904.11263

zbMATH Keywords

softwareordinary differential equationartificial intelligencedifferential equationsditherfixed-point arithmeticIzhikevich neuron modelspinnakerstochastic rounding

Mathematics Subject Classification ID

Numerical methods for initial value problems involving ordinary differential equations (65L05) Numerical algorithms for specific classes of architectures (65Y10)

Related Items (5)

Stochastic Rounding Variance and Probabilistic Bounds: A New Approach ⋮ On stochastic roundoff errors in gradient descent with low-precision computation ⋮ Rethinking arithmetic for deep neural networks ⋮ Numerical algorithms for high-performance computational science ⋮ Stochastic Rounding and Its Probabilistic Backward Error Analysis

Cites Work

• On explicit two-derivative Runge-Kutta methods
• Probabilistic rounding in neural network learning with limited precision
• Shift Register Sequences – A Retrospective Account
• TestU01
• Handbook of Floating-Point Arithmetic
• Accuracy and Stability of Numerical Algorithms
• Lectures on Finite Precision Computations
• Accuracy and Efficiency in Fixed-Point Neural ODE Solvers
• Reprint of a Note on Rounding-Off Errors
• Tapered Floating Point: A New Floating-Point Representation
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Stochastic rounding and reduced-precision fixed-point arithmetic for solving neural ordinary differential equations