Learnings about AI - Week 3's graphic - deep learning is a tool to extend all machine learning. It's really fulfilling to understand what people mean when they say Deep Learning and ANNs (Artificial Neural Networks) via this week's learning in the University of Oxford AI Progamme:
Key Learning: Perceptron is not the new Autobot, it's an electronic simulation of a neuron in the brain. ANNs are made up of ranks and ranks of perceptrons, just like a human brain.
Key Learning: chat-GPT wouldn't work without backpropogation (feeding the results of a learning back into the system to let it self adjust).
Key Learning: Convolutional Neural Networks can process visuals by partitioning them into sections and finding features, then sewing them all up again to work out what is in the picture.
Key Learning: Computers can create visuals by us reverse engineering what we've taught them to recognise. picture+description = classification. So if we give them the classification+description (via text description) = picture.
If you want to freak yourself out a little check out the AI PIN at https://humane.com/ .
