Scientists have grown a human brain, hooked it up to a computer and taught it to solve equations

A new study has demonstrated the promise of combining human brain cells with computers. Live neurons were trained faster than artificial models while showing almost identical results.

Here's What We Know

Using stem cells, researchers were able to create a colony of cells that replicate the structure of neural connections and neurons themselves in the human brain. The colony is called a brain organoid. It was grown on an array of high-density electrodes.

The neural network is called Brainoware. The image below shows it between a week and a few months old. For training, the scientists used more than two hundred audio recordings of Japanese people speaking the main sounds. After training, Brainoware was able to recognise the voice with up to 78% accuracy.

The scientists' next task was to train the neural network to solve Eno mappings. This was achieved in four days. After training, Brainoware was able to solve equations with an accuracy of about 80%. Here, the grown brain performed better than artificial intelligence without a long chain of short-term memory elements (Long short-term memory, LSTM).

If we compare Brainoware and artificial intelligence with LSTM, the AI wins with a slight edge in its favour. However, it is important to note that the artificial neural network went through 50 stages of training, while Brainoware took 10 times less time to train.

Scientists believe that colonies of living neurons will outperform both existing and future neural networks on chips. This applies to both speed and cost-effectiveness. In the case of full-fledged biocomputer systems, ethical concerns are likely to be raised, but this is still a long way off.

Source: Science Alert