The Citadel cadets have emerged victorious in the Amazon Web Services (AWS) DeepRacer Competition, surpassing their rivals from West Point and the Naval Academy. Led by their faculty coach, Dr. Pooya Niksiar, the team consisted of Cadets Frederick Vogel, Blakely Odom, and Brian Bradrick. Through the power of Artificial Intelligence (AI), the Citadel cadets successfully trained AI models to autonomously drive AWS DeepRacer cars and secured first place in the tournament.
Rather than relying on human control, the cadets utilized reinforcement learning, a form of AI, to teach the DeepRacer cars to navigate the track independently. This involved training the cars virtually on a cloud-based simulator and using the acquired AI models to guide the cars during the actual race. The algorithm used was based on deep reinforcement learning, allowing the cars to make decisions based on learned experiences and optimize their performance for long-term goals.
The victory is a testament to the perseverance and dedication of the Citadel cadets, who started from scratch in the summer of 2023. Despite their initial apprehension about participating in an AI-based competition, they embraced the opportunity to explore new technologies and gained valuable insights that apply not only to competitions but also to real-life scenarios. The success of the Citadel team highlights the significance of AI education and its relevance in various fields, such as financial services and autonomous driving.
AWS DeepRacer serves as an engaging platform for participants to delve into reinforcement learning, an advanced machine learning technique. Unlike other approaches, reinforcement learning allows models to learn complex behaviors without the need for labeled training data. This innovative approach empowers participants to make short-term decisions while optimizing for long-term objectives.
The Citadel’s triumph in the AWS DeepRacer Competition exemplifies the institution’s commitment to providing a world-class engineering education. With their Center for AI, Algorithmic Integrity, and Autonomy Innovation (AI3), the Citadel aims to equip cadets, students, and the community with the necessary skills in AI and machine learning. By partnering with AWS Machine Learning University’s AI Educator Enablement Program, the Citadel’s faculty undergo rigorous training to ensure effective instruction in machine learning.
As AI continues to play an integral role in research and industry, the Citadel remains at the forefront of engineering innovation. Their consistent ranking in the Top 25 Undergraduate Engineering programs reflects their dedication to delivering a superior educational experience. With principled leadership and a focus on AI integration, the Citadel continues to shape the future of engineering in South Carolina and beyond.
Q: What is reinforcement learning?
A: Reinforcement learning is an advanced machine learning technique that allows models to learn complex behaviors without relying on labeled training data. It involves providing rewards for good actions and punishing bad actions, similar to training a dog.
Q: How did the Citadel cadets train the DeepRacer car?
A: The cadets trained the DeepRacer car virtually on a cloud-based simulator, using AI models. The car learned from its experiences and trial and error in the simulator until it perfected its navigation skills.
Q: What is the purpose of the AWS DeepRacer Competition?
A: The competition aims to provide an AI teaching tool and promote the use of reinforcement learning. It allows participants to gain hands-on experience with AI technologies and make short-term decisions while optimizing for long-term goals.
Q: What is the significance of the Citadel’s victory in the competition?
A: The Citadel’s victory demonstrates their prowess in AI and reinforces the institution’s commitment to providing a world-class engineering education. It also highlights the potential applications of AI in various industries, including financial services and autonomous driving.