Deep learning without the jargon
Understand what deep learning is, what it does well and where it fails, with no maths needed
A taste of a lesson
If the computer learns by itself, does that mean nobody programmed it?
People still program a lot: they write the code for the network, choose the examples, decide what counts as a right answer and test the result. What they do not write is the rule itself, such as 'a cat has pointy ears'. Instead the network adjusts millions of numbers until its answers match the examples. So it is better to say people program the learning process, and the data shapes the final behaviour. Quick check: if the training photos contained almost no black cats, what might happen when the finished model sees one?
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Explain in plain words what deep learning is and how it differs from ordinary programming
- Describe training and inference using an everyday example
- Name problems deep learning suits well and problems where it is a poor fit
- Spot common misconceptions about AI in news stories and conversations
Lesson plan
- 1 Rules versus examples See why some problems are easier to solve by learning from examples than by writing rules. Start
- 2 What a neural network is, in pictures Picture a network as layers of adjustable dials that turn raw input into an answer. Start
- 3 How training works Understand training as repeated guessing, measuring the error and adjusting slightly. Start
- 4 Why it took off and what it needs Connect the rise of deep learning to data, computing power and better methods. Start
- 5 What it is good at and where it fails Judge which tasks suit deep learning and recognise its typical failure modes. Start
- 6 Reading AI news with a clear head Apply what you learned to judge a real claim about deep learning in the news. Start
Try asking
About this tutor
For curious people with no technical background who keep hearing 'deep learning' and want a clear picture. The lessons start from one everyday example, a program that learns to tell cats from dogs in photos, and build up the ideas of layers, learning from examples and making predictions. You will learn the handful of words that really matter (model, weights, training, inference, dataset), see why deep learning took off when it did, and finish knowing which problems it suits and which it does not. No coding or maths is needed, and every lesson ends with a short check you answer in your own words.
Reviews
4.7
3 ratingsSample
- Tomasz K.Sample
Clear and calm. The news reading lesson was the most useful part for me. I would have liked one more example from medicine, but I understand why it stays general.
- Helen W.Sample
I am retired and wanted to follow what my grandson talks about. The cat photo example carried me through every lesson. I can now explain training versus inference at dinner without getting muddled.
- Amara N.Sample
Finally someone said plainly that it does not understand anything, it learns patterns. That one sentence fixed a lot of confusion I had from articles.
About the teacher
I teach how neural networks learn, one small worked example at a time
9 tutors 424 lessons taught Sample
I teach the core mechanics of deep learning: what a neuron computes, how a loss turns mistakes into numbers, and how gradients and optimisers change weights. My background is in building and training models for applied research teams, which mostly meant staring at loss curves that refused to go down. That shaped how I teach. I start every idea with...
See Mira's profile and tutorsMore like this
Other tutors on the same or nearby topics.