Modeling the Effects of Neurogenesis on Interference and Pattern Separation
WHEN:
January - April 2016
WHAT:
Final Project for Computational
Neuroscience Course
WHERE:
Wellesley College, MA, USA
TOOLS:
MATLAB
January - April 2016
WHAT:
Final Project for Computational
Neuroscience Course
WHERE:
Wellesley College, MA, USA
TOOLS:
MATLAB
Pattern separation is the computational process that minimizes interference of similar memories through the creation of distinct, non-overlapping neural codes. Studies suggest that neurogenesis may influence pattern separation in the dentate gyrus, a structure that provides input to the hippocampus. However, there are scarce investigations into modelling how the role of neurogenesis in pattern separation influences the learning of distally or proximally similar items.
This project serves to replicate the findings of , who were able to computationally model the developmental trajectory of young neurons and their influence on learning and memory. Our model utilizes a simplification of the deep learning algorithm, the Restricted Boltzmann machine.
This project serves to replicate the findings of , who were able to computationally model the developmental trajectory of young neurons and their influence on learning and memory. Our model utilizes a simplification of the deep learning algorithm, the Restricted Boltzmann machine.