- Competition launched in March with AWS experts guiding participants
- Inspiring APU students to explore the full potential of ML and cloud technologies
Imagine racing a self-driving car on a physical track while learning more about Reinforcement Learning (RL), an advanced machine learning (ML) technique. This dream became a reality for racing enthusiasts at the Asia Pacific University of Technology (APU) on April 13th. The university is working with Amazon Web Services (AWS), the most widely adopted cloud service in the world, and he is the first university in Malaysia to host AWS. Deep racer competition.
AWS DeepRacer is an AWS-powered 1/18th scale autonomous race car that is a quick and interesting way to learn about training reinforcement learning models, build self-driving applications to experiment with technology, and test RL models. Designed to provide a way for everyone. Race on a physical track. The event was attended by 215 teams of 535 prospective scientists and engineers studying at APU.
The competition was organized by the university’s Center of Technology and Innovation (CTI), which is responsible for developing technology applications and solutions that will be deployed across APU.
CTI has worked with AWS to move APU’s application architecture to the cloud as part of the university’s ongoing digital transformation strategy. This strategy aims to transform APU’s IT infrastructure into a more agile, efficient and reliable ecosystem that provides a superior learning experience for students, faculty and staff. .
The month-long competition began in March, with AWS experts providing APU attendees with console walkthroughs and hands-on machine learning workshops. Contestants will demonstrate their skills by experimenting with multiple sensor inputs, the latest reinforcement learning algorithms, constructing neural networks, and methods of transfer from simulation to the real domain, before testing their self-driving car in his race. Got a chance to test.
To further hone their technical knowledge and skills, semi-finalists also attended a two-hour coaching clinic conducted by AWS experts to answer technical questions and then submit their race lap times.
The top 10 finalists took part in an exhilarating 2-hour grand finale. Teams ‘The Cloud Champs’, ‘Machine Meowers’ and ‘Common Emitter’ emerged as the top three winners with lap speeds of 8.885, 9.106 and 9.59 seconds respectively.
Julian Lau, Head of ASEAN Growth Markets, Global Public Sector at AWS, joined attendees at the event to talk about their collaboration with APU. “Artificial intelligence (AI) and machine learning (ML) are among the most innovative technologies that help us address some of society’s toughest problems. We are excited to offer students the opportunity to develop their ML skills in an engaging way.Through the AWS DeepRacer League, APU students can explore the full potential of ML and cloud technologies and apply them to real-world scenarios. and to encourage them to apply it to their future careers.”
As part of the APU-AWS DeepRacer Competition 2023, each member of the finalist team will receive 30 hours of simulation training using the AWS console and 5 GB of space to train a reinforcement learning model in a simulated 3D racing environment. Model storage was provided free of charge.
The champions attributed their success to their research and learning from the experiences of others. Shiun said: I am happy that this approach earned him first place. ”
“Machine Meowers” team leader Chong Jack Pang and his team members Tan An Ning and Wong Kai Ze, sophomore second-place winners of Intelligent Systems, learned from the program We credited them for their knowledge-based theory. on the APU. “We learned machine learning and reinforcement learning in the classroom, and that familiarity definitely helped us in this competition.
Team “Common Emitter”, consisting of 2nd year honors students Lee Zhan Tom, Wong Jun Zhi, and Yoong Ming Jun, who have a bachelor’s degree in computer engineering, came in 3rd place after a tough competition. Although they are not machine learning and reinforcement learning specialists, the team is proficient in basic Python coding and has excelled in the machine learning and reinforcement learning process.
“By utilizing a reward and penalty mechanism, we were able to train the machine run by the AI system to perform well. You’ll need to train them in good habits to get there.You’ll be rewarded with high scores for correct moves, but penalized with less points for mistakes, much like the carrot and stick approach. ‘ explained the team.
APU Chief Operating Officer Gurpardeep Singh commended the team for their active participation and dedication to learning, highlighting the value of competition in providing a real-world learning experience. “The APU-AWS DeepRacer Competition 2023 will not only serve as a race, but will also provide practical knowledge and skills to use AI and reinforcement learning to manipulate virtual models and transform these models into real physical models. It provides students with hands-on experience with cutting-edge technology that prepares them for future professional careers.”
The success of the APU-AWS DeepRacer Competition 2023 highlights APU’s commitment to fostering a culture of innovation and providing opportunities for students to excel in new technologies.
