A new transfer-learning framework, Brain2Model, uses human neural recordings to shape the latent representations of artificial networks, improving test accuracy in two tasks.
Semi-orthogonal subspaces for value mediate a binding and generalization trade-off
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.NE 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Brain2Model Transfer: Training sensory and decision models with human neural activity as a teacher
A new transfer-learning framework, Brain2Model, uses human neural recordings to shape the latent representations of artificial networks, improving test accuracy in two tasks.