Loss Functions Compared
Lesson notes
Overview
Squared error punishes large mistakes more than absolute error. This topic is rated intermediate and is number 92 of 117 in Computer Science. Builds on the basics with relationships, methods and trade-offs.
What you are seeing
The panel above shows a graph of the function. Each slider changes one parameter, so you can see exactly how the curve bends, shifts or stretches. The animation is drawn live from the model, so everything you see responds to the sliders. Press Space to pause, use Restart to replay from the start, and change the speed to slow down quick events.
In depth
A loss function measures how wrong a prediction is, and the choice shapes what the model learns. Mean squared error, (ŷ − y)², punishes large mistakes heavily and is smooth to optimise, but is sensitive to outliers. Mean absolute error treats all errors in proportion and is more robust. Cross-entropy is used for classification and strongly penalises confident wrong answers. Choosing the loss means deciding which errors matter most.
Reading the relationship
The expression "MSE vs MAE" is a compact summary of the idea. It uses the symbols M, S, E, v, s, A; name what each one stands for before you start. Then use the sliders to see how changing one quantity changes the others.
A guided exploration
1. Set every slider to its middle value and note the shape of the curve. 2. Move one slider at a time and describe whether the curve shifts, stretches or turns. 3. Find the value where the behaviour changes sharply, then connect it to the formula.
Try this
Predict the outcome first, then test extreme slider values to find where the pattern changes or breaks.
Common pitfalls
Test small inputs by hand and then the largest ones. Off-by-one errors and forgetting how cost grows with input size are the most common traps.
How it connects
It builds on “Softmax Temperature”, the topic before it in Computer Science. Once you are comfortable, continue with “How a Compiler Optimises Code”.
Why it matters
Computing ideas decide how fast, how safely and how well software can solve problems.
Where you might use it
- Revising “Loss Functions Compared” before a class, exam or certification
- Choosing the right algorithm or data structure when writing software or apps
- Debugging programs and judging how fast or how safely code will run at scale
- Understanding how search engines, games, phones and AI systems make decisions
Frequently asked questions
What is Loss Functions Compared?
Squared error punishes large mistakes more than absolute error. It is an intermediate topic in Computer Science.
How do I use the Loss Functions Compared simulation?
The panel at the top shows a graph of the function. Each slider changes one parameter, so you can see exactly how the curve bends, shifts or stretches. Move the sliders one at a time, then press Space to pause the animation and look closer.
What does the formula mean?
The expression MSE vs MAE packs the idea into symbols. Name each symbol first, then change one quantity with the sliders and see which others respond.
Who is this topic for?
Builds on the basics with relationships, methods and trade-offs. If the quiz feels hard, reread the lesson notes and try the previous topic first.
What should I learn before this?
Start with “Softmax Temperature”, the previous topic in Computer Science. Each topic in a subject builds on the ones before it.
How do I track my progress?
Press “Mark as learned” on this page. Progress is stored only in this browser and is never sent anywhere. The Browse page can show topics you have learned or still need to learn.