[ad_1]
Right now, the technology powering the AI chatbots currently on the web is scary, to say the least. It’s immensely powerful, but there was a breakthrough that might leave chatbots like ChatGPT and Google Bard in the dust. This involves neural network technology and how it can improve AI technology.
AI chatbots use neural network technology to deliver their results, but scientists have created a neural network with the ability to make generalizations about language. What does that mean, and why is it significant?
A new neural network is a breakthrough in AI technology
A large part of basic human learning is in adapting new words. When you learn a new word, you’re likely to understand it and induct it into your lexicon to use in your own conversations. This is something that chatbots can do, albeit at an undesirable pace. That doesn’t seem to be a massive issue, as no chatbot you’ve interacted with seemed to be at a loss for words.
However, as Nature reports, scientists have made a breakthrough that might leave current chatbots in the dust. They developed a neural network that uses a new technology called Meta-learning for Compositionality (MLC). With this technology, the neural network is able to learn new words at a much quicker pace than ChatGPT. We’re not only talking about learning the new words but also adding them to its vocabulary for general use.
Along with that, the neural network demonstrated something that could solve one of the major issues facing generative AI today. While testing the neural network, scientists made it so that it would learn from its mistakes. This is huge, as modern chatbots don’t really have the ability.
This means that it could potentially learn from any inaccurate information that it generates. Right now, chatbots struggle with AI hallucinations. This neural network might be the end of those issues. Only time will tell if this is truly the next revolution in AI technology
[ad_2]
Source link