【未来,机器人要有自己的语言了 人类与机器人交流会需要翻译?】

【未来,机器人要有自己的语言了 人类与机器人交流会需要翻译?】美国的一个AI研究团队正在研究如何让机器人创造自己的语言,可以互相聊天。科学家通过“强化学习”来帮助机器人达成这一目标,一旦成功,这对于未来机器人之间互相协作具有重要意义。也许随着这些机器人语言变得更加复杂,人类与机器人进行交谈就需要翻译技术进行支撑。

Robochat: Robots are learning to talk in a completely new language (and humans will need a translator)

  • Robots evolved language to accomplish simple tasks in their 2D world
  • Using trial and error, robots learned which sounds worked and which didn't
  • They could modify their messages to help each other work effectively 

Robots have communicated with each other using a shared language they made up as they went along in an amazing new experiment.

Experts at Elon Musk's artificial intelligence lab have taught robots to start their own language so they could help each other complete simple tasks.

In the future, researchers hope that as the language becomes more complex they will have a 'translator bot' who will be able to convert robot chat into English.

Researchers at OpenAI, which is the artificial intelligence lab created by Tesla founder Elon Musk, have taught robots to use language in order to complete simple tasks (stock image) 

Researchers at OpenAI, which is the artificial intelligence lab created by Tesla founder Elon Musk, have taught robots to use language in order to complete simple tasks (stock image)

WHAT DID THEY DO?

Experts created a two dimensional world to navigate and the robots - or agents - were four coloured circles - green, red and blue.

This layout allowed the robots to communicate with each other in order to complete tasks.

Robots were given different goals such as moving to a specific location or telling others to move somewhere.

They were able to modify their messages to improve the effectiveness of communication.

Agents were able to evolve languages to fit the situation and able to accomplish challenging tasks by communicating with one another.

The team at OpenAI, which is the artificial intelligence lab created by Tesla founder Elon Musk based in San Francisco, created a world where robots learnt a language that enabled them to complete simple tasks - such as moving themselves from A to B.

Experts created a two dimensional world to navigate and the robots - or agents - which were green, red and blue circles had to find their way around.

The robots relied on a technique called reinforcement learning - which involved them trialling different sounds and seeing which ones helped them complete a task and which didn't.

They taught agents to create language by 'dropping them into a set of simple worlds, giving them the ability to communicate, and then giving them goals that can be best achieved by communicating with other agent', the team said in a blog post.

'If they achieve a goal, then they get rewarded.

'We train them using reinforcement learning and, due to careful experiment design, they develop a shared language to help them achieve their goals', they said.

This means that a speaker would for example start to form an association between the word 'house' and the image of a house.

Experts created a two dimensional world to navigate and the robots - or agents - which were green, red and blue circles had to find their way around.

The robots learned a language that enabled them to complete simple tasks - such as moving themselves from A to B. They were given different goals such as moving to a specific location or telling others to move somewhere

HOW DO THEY LEARN?

Robots will use a learning technique called reinforcement learning.

This is basically trail and error - so they keep track of which words work and which don't.

If something works they'll learn to keep doing it.

To build a vocabulary bots will start to connect random sounds with simple concepts.

Locations and objects will start to be linked to certain characters.

As the language becomes more complex experts hope they will be able to translate what they are saying into English.

As they build up their vocabulary they can start using sentences that allow them to convey ideas to one another, writes Wired.com.

As the language becomes more complex, experts hope they will be able to translate what they are saying into English.

The team gave the robots two types of actions - one environment action such as looking at something and communication actions, such as broadcasting a word to one of the other agents.

Robots were able to modify their messages to improve the effectiveness of communication.

For example if an agent realises it could have performed something better had it had supplementary information he will communicate this information to the next agent so its instructions are as clear as possible.

The team found that agents were able to evolve languages to fit the situation and able to accomplish tasks by communicating with one another.

'We hope that this research into growing a language will let us develop machines that have their own language tied to their own lived experience', the researcher wrote.

'We think that if we slowly increase the complexity of their environment, and the range of actions the agents themselves are allowed to take, it's possible they'll create an expressive language which contains concepts beyond the basic verbs and nouns that evolved here.'

http://www.dailymail.co.uk/sciencetech/article-4323464/Robots-learning-talk-completely-new-language.html


Comments are closed.



无觅相关文章插件