Engineering Commons^™

Towards Explainable Ai Using Attribution Methods And Image Segmentation, Garrett J. Rocks

Honors Undergraduate Theses

With artificial intelligence (AI) becoming ubiquitous in a broad range of application domains, the opacity of deep learning models remains an obstacle to adaptation within safety-critical systems. Explainable AI (XAI) aims to build trust in AI systems by revealing important inner mechanisms of what has been treated as a black box by human users. This thesis specifically aims to improve the transparency and trustworthiness of deep learning algorithms by combining attribution methods with image segmentation methods. This thesis has the potential to improve the trust and acceptance of AI systems, leading to more responsible and ethical AI applications. An exploratory …

Project Blipper, Peter Jacobs, Preston Delaware, Ryan Foster

Senior Design Project For Engineers

This project was sponsored by Clorox to design and create an automatic bottle-unscrambling system for possible implementation at their bottling plant in Chile. The objective was to use a robotic arm to unscramble bottles from an incoming conveyor belt and place them upright on an outbound conveyor belt. Throughout the research, design, and testing of solutions for this project, several design alternatives were found for each discipline, and will be presented to Clorox so that they can make an informed decision for how and if they want to move forward with implementation of this project.

The project was split into …

Artificial Neural Network Learning Using Improved Genetic Algorithms, D. K. Chaturvedi

D. K. Chaturvedi Dr.

The feedforward back-propagation artificial neural networks (ANN) are widely used to control the various industrial process, for modelling, simulation of systems and forecasting. The backpropagation learning has various drawbacks such as slowness in learning, stuck in local minima, requies functional derivative of aggregation function and thresholding function to minimize error function. Various researchers have suggested a number of improvement in simple back-propagation learning algorithm developed by Widrow and Holf in 1956. In this paper, a program is developed for feedforward artificial neural network with genetic algorithm (GA) as the learning mechanism to overcome some of the disadvantages of back-propagation learning …