PulseBound: Future-Beat State Forecasting Under an Explicit Information Boundary

Key points

• Predictive representation learning from photoplethysmography (PPG) can violate causal information access even with causal attention, as normalization, nonlocal transforms, or companion views may depend on withheld samples.1
• A shared horizon-conditioned head predicts nine rhythm and morphology descriptors for up to four extractor-valid future beats, using elementwise validity masks; optional ECG-derived pulse-arrival-time supervision is restricted to training.1
• On MIMIC and VitalDB groups held out from PulseBound backbone pretraining, PulseBound reduces nine-state transformed-space MAE relative to last-visible-beat persistence by 28.06% and 22.22%, respectively, with gains in MAE, MAE-Skill, and Spearman correlation across all 40 source-cutoff-horizon cells.1
• These findings separate three testable aspects of predictive physiological representation learning: information access, supervised future structure, and transfer.1

Sources (1)

• [1]PulseBound: Future-Beat State Forecasting Under an Explicit Information Boundary
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 02:10 PM

  We introduce PulseBound, a PPG representation learner combining physiologically structured future-beat prediction with an explicit stored-window information boundary.

  Predictive representation learning from photoplethysmography (PPG) can violate causal information access even with causal attention, as normalization, nonlocal transforms, or companion views may depend on withheld samples.

Extractive summary: sentences quoted from the sources.