Document Type

Honors Project On-Campus Access Only

Abstract

Artificial Intelligence (AI) is an area of computer science that seeks to simulate the ways humans process information and react. Despite being a well-known concept, manifesting itself throughout many facets of our lives, the implementation is a black box. In order to demystify this field, I will cover a crucial branch of AI, deep learning. Deep learning, inspired by the structure and function of the neuron, performs automated feature extraction to enhance learning over many iterations (Brownlee 2019). This was the basis of the research conducted at Macalester College during the summer of 2019 and will supplement the foundational content necessary for understanding neural networks throughout this paper. In doing so, I will illuminate the logic behind the architectural implementations as well as shine light on the projected future of deep learning as more strides are made to increase the robustness and breadth of deep convolutional neural networks. In exploring this field, I learned of the rich history of AI-based research, the complex and unstructured nature underlying the neural network architecture and perfor mance optimization currently, as well as the interesting future routes robotics research conducted at Macalester could potentially take.

Download

Share

COinS
 
 

© Copyright is owned by author of this document