Paper Detail

FlowGenesis: Representation-Driven Self-Organization of Neural Architectures

Maksim A Kazanskii

arxiv Score 10.0

Published 2026-10-26 · First seen 2026-10-27

Research Track A

Abstract

Deep neural networks typically rely on architectures specified before training, despite the continuous evolution of their internal representations. We introduce FlowGenesis, a framework in which neural architectures adapt through neuron birth and pruning guided by Representation Flux, a local statistic capturing changes in minibatch mean activations. Neurons exhibiting persistently high Flux undergo splitting designed to approximately preserve the network function, while neurons with low Flux and low activity are pruned. This mechanism allows network capacity to evolve without prescribing a target architecture or performing a separate architecture search. Experiments on image and text classi- fication show that networks initialized with different sizes approach similar final neuron-count ranges within each dataset while maintaining comparable predictive performance. We further extend FlowGenesis to continual learning, where representation-guided structural adaptation reduces the neurons and parameters allocated to successive tasks while maintaining accuracy comparable to fixed-width Progressive Neural Networks in a task-incremental learning setting. These findings demonstrate that neural architectures can adapt through feedback between representation learning and structural change, suggesting an alternative to prescribing network capacity before training.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{kazanskii2026flowgenesis,
  title = {FlowGenesis: Representation-Driven Self-Organization of Neural Architectures},
  author = {Maksim A Kazanskii},
  year = {2026},
  abstract = {Deep neural networks typically rely on architectures specified before training, despite the continuous evolution of their internal representations. We introduce FlowGenesis, a framework in which neural architectures adapt through neuron birth and pruning guided by Representation Flux, a local statistic capturing changes in minibatch mean activations. Neurons exhibiting persistently high Flux undergo splitting designed to approximately preserve the network function, while neurons with low Flux an},
  url = {https://arxiv.org/abs/2610.35108},
  keywords = {cs.LG},
  eprint = {2610.35108},
  archiveprefix = {arXiv},
}

Metadata

{}