LG Aimers 7th Challenge: A Journey to Becoming an Expert
Image 1. Participants of LG Aimers 7th hackathon
In our daily lives, AI technology is no longer an unfamiliar presence. From self-driving cars and chatbots to search engines, AI is making our lives more efficient across various fields. This advancement is also driving innovation in industries, with LG AI Research at the forefront, leading the use of AI in manufacturing, electronics, and biotechnology.
LG AI Research's "LG Aimers," a practical AI education program for young people aspiring to be AI experts, recently concluded its 7th season. Held over two days from September 6th at the LG Academy in Icheon, Gyeonggi Province, the hackathon was a highly anticipated event, particularly due to its collaboration with LG D&O, a company specializing in space management solutions. Of the 2,570 applicants for LG Aimers 7, 102 participants who had passed a 25:1 competition ratio through online education and a preliminary online hackathon qualified for the final event.
The Mission: Forecasting Demand at Konjiam Resort
Image 2. On-site hackathon of LG Aimers 7th
The mission for the LG Aimers 7th hackathon was to develop a forecast AI model to customer demand for food and beverage outlets at Konjiam Resort. This was a complex forecasting challenge that required participants to consider multiple external factors simultaneously, such as the day of the week, season, and holidays. Given that each outlet had a different customer base, menu, and operational style, participants had to use real data on Konjiam Resort visitors and sales to build a precise demand prediction model. Through this process, they gained practical experience in developing an AI model that could contribute to better food management, efficient staffing, and improved customer satisfaction.
At the offline hackathon, participants worked together to develop the demand forecasting AI model. They discussed and refined various approaches to data analysis, model structure, and learning methods. To achieve accurate predictions, they shared ideas, debated strategies, and continuously monitored a real-time leaderboard. This allowed them to fine-tune their models and fiercely compete for the top spot.
The Challenge: Building the Best Demand Forecasting AI Model
Image 3. Winning teams of the LG Aimers 7th hackathon
- (top) 1st place: Um-nya team, (bottom left) 2nd place: Last Dance team, (bottom right) 3rd place: LGBestModel team
After a fierce two-day competition, the final winners have been determined. Based on a comprehensive evaluation that included leaderboard and presentation scores, the third-place prize went to team 'LGBestModel,' second place to team 'Last Dance', and the grand prize to team 'Um-nya.'
Team 'Um-nya,' the first-place winner, earned high praise for creating a robust baseline optimized for tabular data by combining machine learning and deep learning models. They further enhanced its model by leveraging derivative variables from metadata like resort room and Hwadamsup visitor numbers. The second and third-place teams also received high scores for building forecasting models that considered various data patterns and variables, as well as for making practical suggestions to improve the customer values.
The winning teams shared, "It was a valuable experience and we gained great insights from dealing real-world industry data and building a demand forecasting AI model." They added, "We were able to achieve these results by working together with our teammates."
Connecting for the Future: A Job Fair
Image 4. LG Aimers 7th recruitment fair
On September 7, the second day of the event, a job fair was held with the participation of four LG affiliates: LG AI Research, LG Electronics, LG Energy Solution, and LG CNS. Recruiters provided participants with practical information about employment and offered career counseling. They also informed the participants about registering for the "LG Talent Pool," which gives them priority consideration for future recruitment, thus opening a clear path for them to grow as AI experts.
The End of a Challenge, the Start of a New Beginning
LG provides systematic training and real-world opportunities to cultivate young AI experts. The LG Aimers offers two months of free lectures at a graduate-school level, along with practical training tailored to the hackathon's topic, to help participants enhance their skills. The recent hackathon, in particular, provided specialized training in time-series data analysis, AI model performance optimization, and demand forecasting, preparing participants to tackle complex problems.
With over 17,000 cumulative participants over the past three years, including the 7th season, LG Aimers has established itself as a gateway for "young AI experts." Any young person between the ages of 19 and 29 with basic AI knowledge and coding skills can apply. LG will continue to support the potential of young individuals and provide opportunities to solve real-world industry problems.
LG plans to continue fostering the next generation of talent through LG Aimers, offering them valuable opportunities to solve practical, data-driven problems from various industries, thereby contributing to a bright future for Korea's AI industry.