VarMAE
variable-input masked autoencoder for wrist-worn accelerometry
IMU
Key facts
- Pretraining Data
- NHANES wrist accelerometry (large-scale)
- Venue
- ICLR 2027 submission (under review)
Summary
Two views per signal (long/high-rate and short/low-rate via crop + downsample); masked reconstruction within each view plus a cross-resolution objective (long-view visible tokens predict short-view masked tokens). Matches or beats controlled FM baselines on eight benchmark datasets with one encoder; 0.7M parameters.