Skip to content
Projects

SciML/DeepEquilibriumNetworks.jl

View on GitHub

SciML/DeepEquilibriumNetworks.jl has 59 GitHub stars and 6 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-15 19:34 UTC. Last push 2026-11-03.

How these counts are measured

Measured star growth

in 24h
0
in 7 days
Update delayed
in 15 days
Update delayed
in 1 month
Update delayed

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

About

Implicit Layer Machine Learning via Deep Equilibrium Networks, O(1) backpropagation with accelerated convergence.

Machine Learning

README.md

Original

DeepEquilibriumNetworks

Join the chat at https://julialang.zulipchat.com #sciml-bridged Global Docs

codecov Build Status Build status

ColPrac: Contributor's Guide on Collaborative Practices for Community Packages SciML Code Style

DeepEquilibriumNetworks.jl is a framework built on top of DifferentialEquations.jl and Lux.jl enabling the efficient training and inference for Deep Equilibrium Networks (Infinitely Deep Neural Networks).

Installation

using Pkg
Pkg.add("DeepEquilibriumNetworks")

Quickstart

using DeepEquilibriumNetworks, Lux, Random, NonlinearSolve, Zygote, SciMLSensitivity
# using LuxCUDA, LuxAMDGPU ## Install and Load for GPU Support. See https://lux.csail.mit.edu/dev/manual/gpu_management

seed = 0
rng = Random.default_rng()
Random.seed!(rng, seed)

model = Chain(Dense(2 => 2),
    DeepEquilibriumNetwork(
        Parallel(+, Dense(2 => 2; use_bias=false), Dense(2 => 2; use_bias=false)),
        NewtonRaphson()))

gdev = gpu_device()
cdev = cpu_device()

ps, st = Lux.setup(rng, model) |> gdev
x = rand(rng, Float32, 2, 3) |> gdev
y = rand(rng, Float32, 2, 3) |> gdev

model(x, ps, st)

gs = only(Zygote.gradient(p -> sum(abs2, first(model(x, p, st)) .- y), ps))

Citation

If you are using this project for research or other academic purposes consider citing our paper:

@article{pal2022continuous,
  title={Continuous Deep Equilibrium Models: Training Neural ODEs Faster by Integrating Them to Infinity},
  author={Pal, Avik and Edelman, Alan and Rackauckas, Christopher},
  booktitle={2023 IEEE High Performance Extreme Computing Conference (HPEC)}, 
  year={2023}
}

For specific algorithms, check the respective documentations and cite the corresponding papers.