Imagine eyeing a granite dome at the base of the Cochise Stronghold in southeastern Arizona’s Dragoon Mountains. You’re thinking of doing an all-out granite face climb.
But maybe you should have trained better.
Enter Gabe VanderKlok’s climbing difficulty predictor, SendStone, which the Grand Canyon University senior computer science student developed as part of his yearlong capstone experience, when students create a project that serves as a program exit ticket of sorts before graduation.
VanderKlok showcased his project at Thursday morning’s Artificial Intelligence and Neural Networks Showcase in the campus’s Engineering Building. It is the first AI-specific showcase following the debut this semester of the university’s bachelor’s degree in AI.
He was one of the students in computer science associate professor Dr. Isac Artzi’s AI-neural networks class that explored everything from a handwriting recognition system to an AI tool to respond to customers’ comments on discussion platform Reddit.
Students dabbled in everything from a type of deep-learning algorithm called a convolutional neural network, to an AI model inspired by the human brain called an artificial neural network, to sentiment analyzers.
VanderKlok used all his computer science and AI know-how to create an app for an activity he’s passionate about.
“I love climbing. … I’ve climbed all over the U.S. – Las Vegas, Colorado, Utah, California,” said VanderKlok.
What his project does is tell you how difficult a climb will be.
“There’s no standardized way to predict climbing,” he said. “It’s super subjective. So, it (a particular climb) might be easier for a tall person, a short person … so I thought it would be fun to use AI to predict difficulty.”
He explained that in gyms across the country rock climbers are trained by Kilter Boards, electronic climbing walls with LED-lighted holds that connect to an app. Users can create their own routes to customize their training. Those routes will light up a certain sequence on the board that users can then share with their friends who can try the same sequence in their home gym.
“They have that information stored on a database, so I trained my model on 5,000 different climbs.”
The model is just one of three parts of VanderKlok’s capstone project with fellow teammates that will be presented at the end of the spring.
Fellow senior computer science major Alex Fried also turned to sports for his Head-to-Head NBA Artificial Neural Network, a beautiful blend of analytical nerd-number statistics and the pop culture excitement of professional sports.
His project builds an NBA team based on statistical categories the user wants to focus on, then the user can create a head-to-head matchup with a friend.
“This is mainly for fun and for people that like basketball who want to see, ‘Oh, I can build a better strategy than my friend,’ basically,” Fried said.
“So you’re determining the NBA All-Star team for the Olympics?” said a visitor to the showcase.
“Yeah, yeah! I like that,” Fried said with a smile. “What our strategy should be.”
The visitor suggested using the NBA neural network to help fans build their fantasy basketball teams or for the popular video game series NBA 2K.
From sports to fashion, computer science junior Karik Freiheit built a Chrome extension called Assemblr that allows users to try on clothes virtually.
“Honestly, I thought it was something that I would use,” Freiheit said. “I’ve always been like, why isn’t there a way to try on clothes? Virtual shopping is just so … what? You buy it and if you don’t like it you send it back?”
The Chrome extension – a small software program that adds new features or modifies existing ones to enhance the browser’s functionality – prompts users to upload an image of themselves. The program will show the user how they would look in a certain piece of clothing.
It works on any clothing site, said Freiheit, who completed the Chrome extension in his free time.
Tatum Hansen and Andru Tjalas, both computer science seniors, created a handwriting recognition system “to help people that are new to English learn how to write our alphabet and our numbers,” Tjalas said. “They will write their letters and get immediate feedback.”
Users earn a grade of A through F, depending on how legible the letter is.
Hansen was thrilled when she was asked about whether the system can be used to also teach elementary school students handwriting.
“The original target was to help kids who are in first grade … but it went from a small target to a broader audience,” she said.
Kaden McCarty, a software engineering junior, and Mason Lohnes, a senior computer science major, presented their project, ReplAI, created for a local business as part of their capstone project.
That business wanted an AI tool that would respond to customer comments or posts. At the end of the semester, McCarty and Lohnes’ team will demonstrate what they created to that business.
The AI and Neural Networks Showcase was a pivotal one for the university’s College of Engineering and Technology.
Not only did the college launch the new AI undergraduate program this semester, but it is also offering a new master’s degree in AI. The programs were developed over two years after students expressed an interest in an AI program.
“Many universities are riding the wave of AI,” Artzi said. “ASU just announced this week a Ph.D. in AI.”
For Artzi’s AI-neural network class, he kept the projects open-ended so students could really explore.
“The field of AI is so vast, I didn’t want to put anybody in a box,” he said, adding that one of the things he wanted to do in this showcase is dispel the notion that AI is just ChatGPT.
“So many people, when they hear AI, they think ChatGPT. But you haven’t seen ChatGPT here,” he said, motioning to the students presenting their projects.
“We use a lot of technology that powers the behind-the-scenes of all the big AI companies. Those technologies are freely available to everyone,” he said, and any of GCU’s student AI teams, if they had the same resources, “their projects would be on par with those big AI companies.”
Artzi has said in a past GCU interview, “AI will reshape nearly every field, from medicine and education to how we choose our next Netflix show. My hope is to play the role of translator – helping humans and machines understand each other better.”
GCU News Manger of Internal Communications Lana Sweeten-Shults can be reached at [email protected].
***
Related content:
GCU News: “Teaching A.I.de: How GCU is leveraging artificial intelligence in learning
GCU News: GCU students land internships in AI that track finances, promote ethics
GCU News: GCU conference: AI leads innovations in behavioral health
