So far the model only calculates corner effects, but development continues. People cannot distinguish AI strategies from humans and in many cases choose them.
Investigating the possibilities of artificial intelligence, Google DeepMind laboratory, together with Liverpool football club, developed a TacticAI model that predicts the outcome of events on the football field. The New Scientist website reports this.
Now the TacticAI AI model can predict the outcome of corner kicks and help develop tactics that increase or decrease the likelihood of the final result being a goal. Google DeepMind’s Petar Veličković and colleagues developed the tool as part of a three-year research collaboration with Liverpool. The scientific study was published in the journal Nature.
Important
Corner kicks are awarded when the ball goes out of play over the goal line and can be a good scoring opportunity for the attacking team. According to statistics, about a third of goals are the result of corner kicks or free kicks. That’s why football coaches develop detailed plans for various scenarios that players will learn before the match.
TacticAI used data from 7,176 corner kicks in British Premier League matches from 2020 to 2021, including each player’s position over time, their height and weight. He learned to predict which player would touch the ball first after a corner kick. In testing, the receiver ranked among TacticAI’s top three candidates 78% of the time.
Coaches can use AI to develop offensive or defensive tactics that maximize or minimize the chances of a particular player receiving the ball and the team shooting on goal. The model finds real-life examples of corner kicks for similar patterns and then suggests how tactics can be changed to achieve the desired outcome.
In a blind test, football experts from Liverpool FC (three data scientists, a video analyst and an assistant coach) were unable to distinguish AI-generated tactics from those designed by humans. In 90% of cases, they preferred the outcome of the AI’s work.
Petar Veličković explains the potential threat to coaches: “We strongly support artificial intelligence systems that enhance human abilities and give him more time for the creative part of the job, rather than a system that will replace them.”
The developer says the research could also have broader applications outside of sports. Modeling behavior in football is about modeling human psychology. Using the example of playing football, artificial intelligence is learning how to behave in an open world where events are difficult to predict. He/she will be able to transfer this experience to other situations.
Artificial intelligence is finding more and more applications in the real world, and not all of them are useful for humans. It was recently reported that artificial intelligence produced a clone of a Ukrainian blogger and sold Russian goods in China in his appearance. Ukrainian student Olga Loek, who is studying in the USA and has her own YouTube channel, received a message about this from her subscribers. The clone pretends to be a Russian woman who speaks Chinese, loves China, and wants to marry a Chinese man. The clone is named Natasha, Anna, or Grace, depending on the hosting platform.
Source: Focus
Ben Stock is an accomplished sports journalist. He currently works at 24 news breaker, where he has a reputation for providing in-depth analysis and unique perspectives on the latest sports news. He has a keen eye for detail and a deep understanding of the sports industry. With a passion for storytelling, Ben’s writing takes readers closer to the action and the people behind it.