Stronger inhibitory-to-excitatory synapses in working-memory RNNs reproduce inhibitory dominance, excitatory hypofunction, and task impairment, while resilience training preserves performance at the cost of generalization to longer delays.
arXiv preprint arXiv:2111.01275 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
q-bio.NC 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Derives optimality constraints for nonnegative joint dictionary learning that explain observed SAE behaviors such as feature splitting, absorption, and dense antipodal features.
citing papers explorer
-
Modelling chronic stress as an excitatory-inhibitory perturbation in recurrent working-memory networks
Stronger inhibitory-to-excitatory synapses in working-memory RNNs reproduce inhibitory dominance, excitatory hypofunction, and task impairment, while resilience training preserves performance at the cost of generalization to longer delays.
-
How Optimality Structures Sparse Dictionaries: A Theory for Understanding SAE Representations
Derives optimality constraints for nonnegative joint dictionary learning that explain observed SAE behaviors such as feature splitting, absorption, and dense antipodal features.