Deep learning tools for the measurement of animal behavior in neuroscience
Mackenzie W. Mathis & Alexander Mathis

Pretraining boosts out-of-domain robustness for pose estimation Alexander Mathis, Mert Yüksekgönül, Byron Rogers, Matthias Bethge, Mackenzie W. Mathis

Now Published Preprints:

Using DeepLabCut for 3D markerless pose estimation across species and behaviors Tanmay Nath*, Alexander Mathis*, An Chi Chen, Amir Patel, Matthias Bethge, and Mackenzie W. Mathis. BioRxiv. 2018 (submitted 11/21/2018) *co-first authors

Markerless tracking of user-defined features with deep learning
Alexander Mathis, Pranav Mamidanna, Taiga Abe, Kevin M. Cury, Venkatesh N. Murthy, Mackenzie Mathis*and Matthias Bethge*. aRxiv (submitted 4/8/2018) *co-senior authors |  See our page on DeepLabCut for more info

 Published in Nature Neuroscience:


Gary Kane & Mackenzie W. Mathis. (2019, August 2). Camera Control: record video and system timestamps from Imaging Source USB3 cameras (Version v1.0.0). Zenodo.

Selected past conference contributions:

Applications of deep learning in motor neuroscience, Neural Control of Movement Conference 2019 Team Presentation (talk by Alexander Mathis)

Cerebellar computation 50 years after Marr-Albus: where are we now? Team Presentation (Terry Sanger, Huu Hoang, Mitsuo Kawato, Kazuo Kitamura, Mackenzie Mathis) Neural Control of Movement Conference 2019

Talk during the workshop on Quantifying Social Behavior: Computational challenges and experimental pitfalls. Organized by Annegret Falkner & Ann Kennedy. COSYNE 2019

DeepLabCut: a tool for fast, robust, and efficient 3D pose estimation. Alexander Mathis, Tanmay Nath, Richard A. Warren, An Chi Chen, Amir Patel, Venkatesh N Murthy, Mackenzie Mathis*, Matthias Bethge*. COSYNE 2019

Markerless tracking of user-defined anatomical features with deep learning
Mackenzie W. Mathis*Alexander Mathis*Pranav MamidannaTaiga AbeVenkatesh N. MurthyMatthias Bethge 
May 2018, Neural Control of Movement Conference, Santa Fe

Elucidating the role of sensorimotor cortex for motor adaptation using hierarchical optimal feedback control theory
Travis DeWolfMackenzie W. MathisAlexander Mathis 
May 2018, Neural Control of Movement Conference, Santa Fe

Motor adaptation in mice is driven by task-relevant error-based signals, and dependent on intact cerebellar outputs
Alexander MathisTaiga AbeMackenzie W. Mathis & Naoshige Uchida 
May 2017, Neural Control of Movement Conference, Dublin (currently unpublished elsewhere)

Somatosensory Cortex Plays an Essential Role in Forelimb Motor Adaptation in Mice
Mackenzie W. MathisAlexander MathisNaoshige Uchida 
February 2017, Cosyne, Salt Lake City
May 2017, Neural Control of Movement Conference, Dublin & COSYNE 2017 (appeared in Neuron)