# Relaxed activation analysis of dataflow networks - A clock calculus for machine learning and real-time scheduling

arXiv: https://arxiv.org/abs/2607.21797 · HTML: https://mlsanity.com/paper/2607.21797/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 35 references that are in the map.
- First posted on arXiv 23 Jul 2026.

## Builds on (most cited first, 5 of 5)

- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Streaming keyword spotting on mobile devices](https://mlsanity.com/paper/2005.06720/) (2020)
- [Silent Bugs in Deep Learning Frameworks: An Empirical Study of Keras and TensorFlow](https://mlsanity.com/paper/2112.13314/) (2021)
- [Dynamic Control Flow in Large-Scale Machine Learning](https://mlsanity.com/paper/1805.01772/) (2018)

## Abstract (arXiv)

Previous work has shown that the simple dataflow primitives of the Lustre language allow the natural, semantically unambiguous, and compact representation of machine learning (ML) applications, including models featuring complex conditional execution and recurrent state. The Lustre clock calculus is responsible for the static determination of important properties such as liveness (absence of deadlocks) and static memory bounds. Yet existing clock calculi are tailored for embedded control applications. We show they do not cater for the representation of control patterns commonly found in training algorithms, resulting in cumbersome expressions and inefficient compilation. We propose a conservative extension of Lustre's clock calculus addressing this limitation, thereby facilitating the embedding of ML models in reactive applications.