BDO frames fMRI representation learning as stochastic optimal control of a continuous latent state process, reporting strong downstream performance with fewer parameters than BrainLM and BrainJEPA.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Foundational Brain Dynamics Model via Stochastic Optimal Control
BDO frames fMRI representation learning as stochastic optimal control of a continuous latent state process, reporting strong downstream performance with fewer parameters than BrainLM and BrainJEPA.