Dynamical Inference Lab

Welcome 👋

This is the Github account of the Dynamical Inference Lab at Helmholtz Munich.

We build machine learning algorithms for representation learning and inference of nonlinear system dynamics. We apply these algorithms to model complex biological systems in neuroscience, cell biology and other life science applications.

If you are interested in representation learning algorithms and/or their applications, please reach out! We are hiring for fully-funded PhD, postdoc and research engineering positions, and also welcome undergraduate and Master’s students to apply for research visits, internships and research assistant positions.

Further information on the lab website: dynamical-inference.ai

Popular repositories Loading

  1. dcl dcl Public

    Official reference implementation of dynamics contrastive learning (DCL).

    Python 30 5

  2. Implementation of PatchSAE as presented in "Sparse autoencoders reveal selective remapping of visual concepts during adaptation"

    Jupyter Notebook 30 5

  3. Official implementation of CytoSAE: Interpretable Cell Embeddings for Hematology

    Jupyter Notebook 23 3

  4. FastSAE: Adopt seamlessly, train fast

    Python 4

  5. 🐟 ituna – tune machine learning models for empirical identifiability and consistency

    Python 4

  6. ebc ebc Public

    Official implementation of Equivariance by Contract (EbC)

    Python 1 1

Repositories

Showing 9 of 9 repositories

  • ituna Public

    🐟 ituna – tune machine learning models for empirical identifiability and consistency

    dynamical-inference/ituna’s past year of commit activity

    Python

    4

    MIT 0

    0 1

    Updated Feb 27, 2026

  • dynamical-inference/tdmpc2’s past year of commit activity

    Python 0 MIT

    177 0 0

    Updated Jan 30, 2026

  • ebc Public

    Official implementation of Equivariance by Contract (EbC)

    dynamical-inference/ebc’s past year of commit activity

    Python

    1 1 0 0

    Updated Dec 5, 2025

  • fastsae Public

    FastSAE: Adopt seamlessly, train fast

    dynamical-inference/fastsae’s past year of commit activity

    Python

    4

    MIT 0

    0 0

    Updated Nov 19, 2025

  • patchsae Public

    Implementation of PatchSAE as presented in "Sparse autoencoders reveal selective remapping of visual concepts during adaptation"

    dynamical-inference/patchsae’s past year of commit activity

    Jupyter Notebook

    30

    MIT

    5 0 0

    Updated Oct 31, 2025

  • dynamical-inference/paperbee’s past year of commit activity

    Python 0 MIT

    14 0 1

    Updated Sep 18, 2025

  • cytosae Public

    Official implementation of CytoSAE: Interpretable Cell Embeddings for Hematology

    dynamical-inference/cytosae’s past year of commit activity

    Jupyter Notebook

    23 3 1 0

    Updated Jul 17, 2025

  • dcl Public

    Official reference implementation of dynamics contrastive learning (DCL).

    dynamical-inference/dcl’s past year of commit activity

    Python

    30

    Apache-2.0

    5 0 0

    Updated Jun 24, 2025

  • dynamical-inference/.github’s past year of commit activity

    0 0

    0 0

    Updated Sep 16, 2024