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.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Defines the walk-length filtration for persistent homology on directed graphs, establishes stability under a generalized L1-style network distance, supplies a computation algorithm, and compares it to the Dowker filtration on cycle and synthetic hippocampal networks.
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.
-
The Walk-Length Filtration for Persistent Homology on Weighted Directed Graphs
Defines the walk-length filtration for persistent homology on directed graphs, establishes stability under a generalized L1-style network distance, supplies a computation algorithm, and compares it to the Dowker filtration on cycle and synthetic hippocampal networks.