Graduation Year

2023

Document Type

Thesis

Degree

M.S.C.S.

Degree Name

MS in Computer Science (M.S.C.S.)

Degree Granting Department

Computer Science and Engineering

Major Professor

Alfredo Weitzenfeld, Ph.D.

Committee Member

Sudeep Sarkar, Ph.D.

Committee Member

John Murray-Bruce, Ph.D.

Committee Member

Shaun Canavan, Ph.D.

Keywords

Information Encoding, LSTM, Navigation, Place Cells, Transfer Learning

Abstract

The human brain still has many mysteries and one of them is how it encodes information. The following study intends to unravel at least one such mechanism. For this it will be demonstrated how a set of specialized neurons may use spatial and temporal information to encode information. These neurons, called Place Cells, become active when the animal enters a place in the environment, allowing it to build a cognitive map of the environment. In a recent paper by Scleidorovich et al. in 2022, it was demonstrated that it was possible to differentiate between two sequences of activations of a grid of Place Cell neurons that differ by a speed encoding following the same path. This effect was first perceived using trained rats and then reproduced using a mathematical model of place cells in a robot. With the above result, we used a recurrent neural network to analyze the sequence of activations and produce a decision output. First it used a speed encoding of just 2 different outputs. Given the results achieved with that type of encoding, it was later used a different type of encoding with 6 different outputs. A robot simulation environment was used to test the performance of the model when transferred to a robot controller.

Download

Share

COinS