Today we had an interesting class on LLMs (Large Language Model). Since GenAI and LLMs are the latest trends, thought I will write a bit on this. The class was very informative. Key point I want to highlight is robots are not going to take over the world or anything. That happens only in movies. You would know that if you know that artificial intelligence is just a bunch of algorithms. And they are not going to replace jobs which require human intelligence.
The history of LLMs starts with GAN (Generative Adversarial Network) and VAE (Variational Autoencoder). Then we have transformers which is an encoder-decoder model. Here the ML model can understand the context. Then we have GPT models. GPT models are mostly text based. We need to convert text to vector representation that the computer understands. And next we have DALL-E and stable diffusion model which is generating images from text description. Neural networks are at the base of all these models. Here is my note from the class taken as a mind map.
