Max Xu
@maxxu05
Followers
57
Following
151
Media
2
Statuses
23
Joined June 2020
Only 10 days left to submit to the Time-series 4 Health Workshop at #NeurIPS2025 ! Call for Papers: https://t.co/8ZfrTgjl2M
1
5
15
This year, submissions to the proceedings track will have the option to cross-submit to the #NeurIPS2025 Learning from Time Series for Health Workshop ( https://t.co/3OaCjm5Q1o). More details here: https://t.co/lhv8812Qge
#ML4H
🩺📈 The "Learning from Time-series for Health (TS4H)" workshop is BACK at NeurIPS 2025 🥳! This workshop unites researchers across health time-series domains (from wearables to clinical systems) to tackle shared challenges. Details: https://t.co/X2pw6WTX9h 🧵 (1/6)
0
1
10
this is my recent research at @Google on learning #foundationmodels for #wearable sensors in real-world settings with real-world missingness and noise!!!
Introducing LSM-2, our newest foundation model for wearable sensor data. LSM-2 uses Adaptive & Inherited Masking, a novel self-supervised framework, to learn from incomplete data & achieve strong performance without requiring explicit imputation. More → https://t.co/jeMvzVupZg
0
0
2
Introducing LSM-2, our newest foundation model for wearable sensor data. LSM-2 uses Adaptive & Inherited Masking, a novel self-supervised framework, to learn from incomplete data & achieve strong performance without requiring explicit imputation. More → https://t.co/jeMvzVupZg
9
63
364
If you are working on AI in healthcare, please consider submitting to our Time-series 4 Health workshop at NeurIPS 2025!! We have prepared a great lineup of speakers and talks ⌚️🏥⏱️📈🧠🫀💃
🩺📈 The "Learning from Time-series for Health (TS4H)" workshop is BACK at NeurIPS 2025 🥳! This workshop unites researchers across health time-series domains (from wearables to clinical systems) to tackle shared challenges. Details: https://t.co/X2pw6WTX9h 🧵 (1/6)
0
0
3
🩺📈 The "Learning from Time-series for Health (TS4H)" workshop is BACK at NeurIPS 2025 🥳! This workshop unites researchers across health time-series domains (from wearables to clinical systems) to tackle shared challenges. Details: https://t.co/X2pw6WTX9h 🧵 (1/6)
1
7
17
Excited to be in beautiful Singapore for #ICLR2025! Come see our poster on RelCon, SOTA foundation model for motion data, led by stellar @maxxu05 with Apple Health Research. Poster 17 in session 6 (3:00 - 5:30 April 26 Hall 3+2B)
arxiv.org
We present RelCon, a novel self-supervised Relative Contrastive learning approach for training a motion foundation model from wearable accelerometry sensors. First, a learnable distance measure is...
0
3
15
Thank you so much to all of my amazing collaborators at Apple (Jaya, Greg, Haraldur, Shirley, @HyewonMandyJ, and many more I can't list) and my amazing PI, @RehgJim, for all of their endless help and support!!
0
0
1
My paper RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable Data, from my @Apple internship, has been accepted at #ICLR2025! 🎉 We introduce the first IMU foundation model, unlocking generalization across motion tasks. 🏃♀️📊 https://t.co/BGVzFjSjxn
arxiv.org
We present RelCon, a novel self-supervised Relative Contrastive learning approach for training a motion foundation model from wearable accelerometry sensors. First, a learnable distance measure is...
1
10
17
Poster presentation alert 🚨 I will present the paper “Event-Based Contrastive Learning” in Poster Session B (Poster #156, 3:35-4:55PM) @mlforhc ! Link to camera-ready version of paper: https://t.co/IrqqcEer80
🌟 Excited to share our latest work "Event-Based Contrastive Learning for Medical Time Series" now available on Arxiv! 📄 https://t.co/91wXq3DSV4. A big shout out to my amazing co-authors: Nassim, @MattBMcDermott @AparnaBee @payal_chandak @MarzyehGhassemi, and Collin! 🌟
2
6
32
#ICLR2024 How can we choose meaningful positive pairs for time-series contrastive learning? What about motif similarity? REBAR uses a learned measure that captures motif similarity and achieves SOTA performance. Arxiv: https://t.co/hvYfgZD1wO Github: https://t.co/e5oNccvzlz
1
10
15
How can we fill in missing pulsative sensor data? Prior state-of-the-art fails in our novel setting, despite its well-defined temporal structure. Checkout our #NeurIPS2022 paper, PulseImpute, @ 4 pm CST! arxiv: https://t.co/Hbv2x7ZvkP github: https://t.co/bTManuLyEH
0
11
17
Dense self-supervised learning from multiple 3D viewpoints → dense feature representations that generalize both to novel object instances and to novel categories of instances. Checkout our #NeurIPS2022 paper! arxiv: https://t.co/rxzdrScII1 github: https://t.co/5n98W7Wykt
2
20
68
the prophet speaks
is joe gonna run for president?? you could say he's Biden his time
0
0
0
Congrats to my Ph.D. students Fiona Ryan (@fionakryan) and Max Xu (@maxxu05) for both winning NSF Graduate Fellowships. 🎉🎉🎉 #ProudAdvisor @mlatgt @ICatGT
0
6
31
Hmmm
0
0
1