How Much is Brain Data Worth for Machine Learning?
For researchers in NeuroAI and machine learning, this work provides a theoretical framework to quantify the value of neural recordings for improving model training, though it is an incremental theoretical contribution.
This paper theoretically analyzes when and how much brain data can improve machine learning performance, deriving scaling laws and exchange rates between brain and task samples. It finds that brain data can be valuable under certain conditions of task-brain alignment and noise, but the benefit is modest and context-dependent.
If a person can solve a task, can measuring their brain make it easier to train a model to solve that task too? Recent NeuroAI work suggests that supplementing task training with neural recordings can modestly improve model performance and robustness. However, it is unclear when there should be a benefit from using neural data and how much benefit to expect. We formulate this question mathematically, and begin to address it theoretically using a simple, analytically tractable linear gaussian model of task targets and neural recordings. For a multimodal estimator trained on both brain data and task labels, we derive scaling laws for how performance scales with the numbers of brain and task samples. From these laws we derive relative value and exchange rates between brain samples and task samples, quantifying how much extra task samples neural data is worth as a function of task-brain alignment, neural and task noise, latent dimension, and brain data sample size. We also analyze test distribution shift, to identify conditions where brain-regularized learning can produce substantial robustness gains through learned invariances. Finally, under a fixed collection budget, we characterize the regimes in which brain data is worth collecting. Our results provide a foundation for understanding how valuable brain data could be for improving machine learning.