How Does AI Learn? A Beginner’s Guide with Examples – BlueDot Impact
This article explains how AI systems learn parameters from data, and how advances in compute, data and algorithms have enabled modern AI systems. Learning parameters from data Imagine you’re an estate agent, and want to know how to price houses you’re selling. You have some basic data about previous sale prices, which looks like this: […]
This article explains how AI systems learn parameters from data, and how advances in compute, data and algorithms have enabled modern AI systems. Imagine you’re an estate agent, and want to know how to price houses you’re selling. You have some basic data about previous sale prices, which looks like this: What you might do to evaluate this is draw a line of best fit through the graph. This has the equation y = ax + b, where a and b are parameters. To find the parameters, you can use trial and improvement.[[1]] Perhaps you start by picking some numbers at random for our parameters, and see…
saved by
related reading
- CSET-AI-Triad-Report.pdfcset.georgetown.edu
- The Little Book of Deep Learningfleuret.org
- Gradient descent, how neural networks learn | 3Blue1Brown3blue1brown.com
- Neural networks and deep learningneuralnetworksanddeeplearning.com
- Supplement to “Why AI alignment could be hard”cold-takes.com
- There Are No New Ideas in AI… Only New Datasetsblog.jxmo.io
- Andrej Karpathy — AGI is still a decade awaydwarkesh.com
- The Only Important Technology Is The Internet - Kevin Lukevinlu.ai
- Just Ask for Generalization | Eric Jangevjang.com
- NL.pdfabehrouz.github.io
- GenAI Handbookgenai-handbook.github.io
- Discovering 108 tricks to accelerate grokkingkindxiaoming.github.io