In 1950, Alan Turing introduced the Imitation Game, now known as the Turing Test, to evaluate computer intelligence. Generative AI has passed this test, leading to the pursuit of Artificial General Intelligence (AGI), an advanced form of AI capable of human-level understanding and learning. While AGI remains undefined and unrealized, researchers in China propose the Survival Game as a new evaluative tool based on trial and error, determining AI capability in various tasks including image classification. This method allows AI models to 'survive' and progress based on their performance.
"The main idea behind this paper is to assess whether current AI systems can find solutions through continuous trial and error," Jingtao Zhan, a PhD student in computer science at Tsinghua University and corresponding author, told The Register.
"If an AI system can find a solution within a limited number of attempts, it is considered to 'survive'; otherwise, it 'goes extinct.'"
Models that survive are allowed to progress to other tests; ones that don't pass get retrained until they do, which is a significant process.
The Survival Game covers various knowledge domains. In image classification, for example, the test assesses how many trial-and-error attempts are required before.
Collection
[
|
...
]