Exact Network Surgery and Reactive Computational Graphs in Julia with NeuroDSL Hi everyone, I’d like to share some recent theoretical and systems results from NeuroDSL, a persistent, reactive computational graph framework for Deep Learning built entirely in Julia. While tape-based and define-by-ru…

Thanks, you’re seemingly doing great work. I want to understand this more, so far scanning the papers. Conclusion in paper ends with: Together, these results turn “growing a network mid-training” from a fragile engineering trick into an operation with a specification, a proof, and a passing test …

@Palli Thanks for taking the time to read through the papers so carefully! You’ve highlighted some of the exact intersections between theory and systems engineering that motivated this project. To address your points: 1. Compatibility with KeemenaLM.jl (and Flux.jl models) You are completely righ…

Hi! Thanks so much for reaching out and for your interest in the project! The short answer is: not directly “plug-and-play” out of the box, but yes, it is absolutely possible with a small conversion step. Here is why: KeemenaLM.jl is built on top of Flux.jl, which uses a standard define-by-run app…

If a Lux.jl backend was used in KeemenaLM.jl would it be possible to have them work together? That has not be implemented but the design does allow for different backends to be used if implemented.

[image] Khemais_Abdallah: Hi! Thanks so much for reaching out and for your interest in the project! Thank you very much for the detailed reply and for introducing me to NeuroDSL. It looks like a very interesting package! Also, many thanks to @Palli for the additional information. While I can…

Thank you for the question. As you noted, it hasn’t been implemented yet, but the separation of concerns in the graph engine’s design makes that kind of interoperability entirely feasible in principle. NeuroDSL treats the computational graph as a persistent DAG where nodes own their cached values a…

Really insightful, can you expand into more technical details on what would be needed for the state handling? On a high level I grasp what you are saying.

To get into the technical weeds: the friction (and the solution) comes from bridging a purely functional paradigm with a persistent stateful DAG. In Lux.jl, a forward pass is strictly explicit: y, st_new = Lux.apply(layer, x, ps, st). The layer itself holds no state; parameters (ps) and states (st,…

Are you planning to update NeuroDSL in the near future by adding LuxNode wrapper to make it fully compatible with Lux, or is that currently rather outside your scope of interest?

Exact Network Surgery and Reactive Computational Graphs in Julia with NeuroDSL

Specific Domains Machine Learning

j_u July 18, 2026, 8:14pm 2

Hi, can your package be used with KeemenaLM.jl?

Topic		Replies	Views
Mutable Computational Graphs in Pure Julia: Performing "Hot Surgery" on a LLaMA Transformer Machine Learning	0	178	July 8, 2026
[ANN] Lux.jl: Explicitly Parameterized Neural Networks in Julia Package Announcements package , announcement , machine-learning	49	12247	April 16, 2024
Deep learning in Julia Machine Learning	35	13292	April 22, 2024
State of machine learning in Julia Machine Learning	60	67275	August 26, 2022
Speed Comparison Python v Julia for custom layers Machine Learning question , flux , python , machine-learning , speed-optimization	25	1775	July 28, 2023

Exact Network Surgery and Reactive Computational Graphs in Julia with NeuroDSL

Related topics