AI with ‘spontaneous’ behaviour like animals developed
The study could be a step towards developing synthetic brains in artificial bodies
Your support helps us to tell the story
From reproductive rights to climate change to Big Tech, The Independent is on the ground when the story is developing. Whether it's investigating the financials of Elon Musk's pro-Trump PAC or producing our latest documentary, 'The A Word', which shines a light on the American women fighting for reproductive rights, we know how important it is to parse out the facts from the messaging.
At such a critical moment in US history, we need reporters on the ground. Your donation allows us to keep sending journalists to speak to both sides of the story.
The Independent is trusted by Americans across the entire political spectrum. And unlike many other quality news outlets, we choose not to lock Americans out of our reporting and analysis with paywalls. We believe quality journalism should be available to everyone, paid for by those who can afford it.
Your support makes all the difference.Scientists are developing artificial intelligence that is able to recreate the spontaneous activities of animals.
Researchers at the University of Tokyo in Japan used a neural network – a computer program that mimics the brain – to try and make robots change their behaviour autonomously, as if it was based on instinctive changes.
“Basically, our work is about how to design spontaneous behavior switching by exploiting chaotic dynamics,” said co-author Kohei Nakajima, an applied mathematician at the university.
These chaotic dynamics refer to when a system quickly, but deterministically, switches between stereotypical patterns – such as quickly changing actions between walking and running.
Roboticists have attempted to achieve this before, by simulating a system with hundreds of motors that represent muscles, and each one was connected to a chaotic oscillator which acted like motor neurons.
The system could then reproduce motions similar to these chaotic actions by switching between stereotypical behaviour, such as crawling or rolling.
Following that, scientists attempted to develop a system that used a high-level neural network to control lower-level activities, but the time it takes to develop the algorithm is very long. As such, this new study does not use a hierarchical design but rather a three-step method using a machine learning framework.
The researchers defined several possible behaviours, and trained the neural network to act on them according to commands. The network would then switch between those behaviours in a specific order, and then finally a system was devised that would transition between the behaviours chaotically.
"The final goal is to somehow realize animal-like actions, and animals have spontaneity,” Nakajima said.
The study, however, was limited to only working on a computer, rather than in a physical robot.
“The true challenge is now on its embodiment and if they can obtain similar results,” said computer scientist Alexandre Pitti at CY Cergy Paris University who was not related to the study, adding that the research may be a step towards “a synthetic brain that can have memory that can interact with the environment through an artificial body.”
Join our commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies
Comments