Alexander Mathis
@TrackingPlumes
Followers
4K
Following
3K
Media
58
Statuses
1K
Assistant Professor in Comp Neuro and ML at @epfl_en @CampusBiotech | #deeplabcut co-developer | past: @LMU_Muenchen, @uni_tue and @Harvard
Geneva, CH
Joined April 2015
This summer I spent time in Alaska doing field work with bears. It was a wonderful experience to learn more about bears and meet our collaborators!!
7
1
80
Just in time for tracking my drills over the holidays. .
What a year! #DLC3🔥 is up tested by many of you (👀 watch out for a paper in 2025), we hit 750K downloads, and have over 120 contributors to the code! . ❄️⛷just dropped: new human pose models & high perf. models! #DigvijaySinghRathee . HAPPY HOLIDAYS!
1
0
21
RT @DeepLabCut: What a year! #DLC3🔥 is up tested by many of you (👀 watch out for a paper in 2025), we hit 750K downloads, and have over 12….
0
13
0
RT @epflSV: 🌟 Join the 2nd Lemanic Life Sciences Hackathon! 🌟.Are you a clinician with questions data could answer?.A life scientist with d….
0
2
0
RT @NeuroCellPress: Here's our latest issue: The cover is for @TrackingPlumes & coll.'s work on a novel learning al….
0
17
0
I'm humbled to receive this award with @SuseWegener and @TrackingActions - thanks to all my students, collaborators and former supervisors that made this possible! . Let's keep pushing to better quantify behavior for health and beyond!.
🥳🍾Alexander Mathis @TrackingPlumes and Mackenzie Mathis @TrackingActions the 2024 Robert Bing Prize, with Susanne Wegener @SuseWegener 🏆.
3
0
50
RT @DeepLabCut: 🚀 DLC AI Residents call for 2025! This year, we are doing something a little different to support even more underrepresente….
0
12
0
RT @amathislab: Earlier this year we got a new lab logo -- it illustrates the core questions we are interested in. Measuring behavior, an….
0
4
0
RT @TrackingActions: ✨🥰 check out our article - and cover 🤩- about Decoding the Brain in @CellCellPress . We revie….
0
76
0
RT @SebastianSeung: But is it physics? Hopfield tackled three profound questions about emergence: condensed matter from interactions betwee….
0
46
0
Truly foundational!.
BREAKING NEWS.The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”
0
0
11
RT @SebastianSeung: A historic milestone for neuroscience! It's been a fantastic journey for the #FlyWire Consortium. This special edition….
0
91
0
Check out our work on the unsupervised discovery of hierarchical behavior from motion capture data at #ECCV24 this afternoon!.
I'm thrilled to present our paper "Elucidating the Hierarchical Nature of Behavior with Masked Autoencoders" at #ECCV24! 🎉 We focus on self-supervised learning for action segmentation to uncover the hierarchical structure of behavior. Paper: 👇🧵.
0
1
18
The 2024 edition is ongoing, so feel free to join the competition. .We'll sit this one out, ;).
🚨#MyoChallenge Submissions are LIVE.MyoChallengers, you can now register your team on evalAI and submit your best solution and start competing against worldwide teams 🌏.More info ➡️ May the best teams win 🏅
0
0
2
Fun fact, we also won the #MyoChallenge at @NeurIPSConf in 2023, also using curriculum learning & Lattice:
1
0
5
Overall, our work illustrates the emerging possibilities at the interface of musculoskeletal physics engines, reinforcement learning and neuroscience to advance our understanding of biological motor control. Join the grand challenge!.
1
0
5
This collaboration started at a 2022 BBQ 🍖 🤗 ➡️ Alberto, Pablo, & Nisheet worked tirelessly to win the challenge. In the next summer, Alberto with Abigail, who was then an undergraduate in the lab, analyzed the policy. It was a great journey!.
1
0
6
By selectively inactivating parts of the control signal, we found that more dimensions contribute to the task performance than suggested by traditional synergy analysis. The learned muscle synergies are highly task-specific & thus generalize poorly to other tasks.
1
0
5
The synergy analysis is based on common reconstruction methods. We argue that task performance is a key metric beyond signal reconstruction.
1
0
4
In agreement with data from human subjects, the policy uncovers a small number of kinematic synergies even though it is not explicitly biased towards low-dimensional solutions!
1
0
3