A single recurrent network trained on masked sensory prediction and motion develops co-emergent grid and place cells that qualitatively match multiple experimental observations without any spatial supervision.
The internal wall is removed, and the agent is allowed to move freely through the full arena
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-bio.NC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A simple model of co-emergence of grid and place fields
A single recurrent network trained on masked sensory prediction and motion develops co-emergent grid and place cells that qualitatively match multiple experimental observations without any spatial supervision.