The machine that loved Rachmaninoff

Everything you're seeing in the AI space is driven by the new wonders of GenAI, but there's more to AI than LLMs and there's ways to apply it that doesn't need huge scale to work. In this talk, I'll present a hobby project in which I used neural networks to create a simple language model for music and taught an AI Bot to appreciate the same music as my son. Why this? 1. It's different, interesting (IMO) and it's innovative 2. It has real world application in the field of recommendation engines 3. It proves that you can do a lot with even the tiniest of neural networks (less than 100 virtual neurons) What you'll take away 1. An appreciation of how AI works at the fundamental level 2. An awareness that AI is something individuals can develop on a shoestring budget and not just the preserve of multi-billion dollar funded organisations 3. An awareness that there's more to AI than LLMs