{
  "abstract": "Simplified spiking neuron models such as the Adaptive Exponential Integrate-and-Fire (AdEx) model reproduce the firing behaviour of diverse neuron types at a fraction of the cost of biophysically detailed models, making them attractive for large-scale brain simulation. Fitting their parameters to recordings, however, still relies on derivative-free search such as grid search or evolutionary algorithms, because the discrete spike-and-reset mechanism renders these models non-differentiable. Surrogate gradients, used in deep spiking networks, replace the spike derivative with a smooth approximation on the backward pass. Whether they also enable efficient gradient-based parameter estimation for single simplified neurons is untested. We implement a differentiable AdEx model with a surrogate-gradient variant in the Jaxley framework. The implementation is on par with target-compiled C++ Brian2 code while remaining end-to-end differentiable, vmap-batchable and GPU/TPU-portable. It is 200 times faster than Brian2's default runtime mode. Across a recovery-radius benchmark of 2,700 paired runs, gradient-based fitting never matched a gradient-free Nelder-Mead baseline. At sub- threshold levels, gradient descent instead beats Nelder-Mead. The obstacle is therefore the spike mechanism and the loss geometry it induces, not gradient descent itself.",
  "abstract_shown": true,
  "affiliations": [
    "KTH Royal Institute of Technology"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 2,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "P. Mayer"
    },
    {
      "name": "A. Kozlov"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/7x1ytnw3p27ew.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.09.19.752831",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.09.19.752831v1.full",
  "id": "7x1ytnw3p27ew",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.09.19.752831v1",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ],
    [
      "openalex",
      "2805|Neural dynamics and brain function"
    ]
  ],
  "openalex": "W7214353016",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.09.19.752831v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-19",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/7x1ytnw3p27ew.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:24:28Z",
      "native_id": "10.64898/2026.09.19.752831",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.09.19.752831v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Surrogate Gradients for Gradient-Based Parameter Estimation in Simplified Neuron Models",
  "title_text": "Surrogate Gradients for Gradient-Based Parameter Estimation in Simplified Neuron Models",
  "type": "preprint",
  "updated_at": "2026-10-30T12:24:28Z",
  "url": "https://research.draconitic.com/paper/7x1ytnw3p27ew",
  "venue": "bioRxiv",
  "versions": []
}