Skip to content

Neural Networks: First Intuition

Why this matters

The full neural-network lesson uses terms like weights, bias, loss, gradient, learning rate, forward propagation, backpropagation, and chain rule.

This page gives the basic idea first:

make a prediction -> measure how wrong it was -> adjust the knobs -> try again

You do not need calculus, matrices, or Python code yet. Start with the learning loop.

Mental model

Imagine a cafe wants to predict whether an order will be small or large.

order items ordered drink size includes food real answer
A 1 small no small
B 3 large yes large
C 2 medium yes large
D 1 large no small

The model starts with a rough rule. It might care too much about drink size and not enough about whether food is included.

Training means improving those internal settings after seeing examples.

Plain version:

bad setting -> wrong predictions
better setting -> fewer wrong predictions

Core ideas

  • A feature is one input fact the model can use.
  • A prediction is the model's answer for one example.
  • Weights are adjustable settings that control how strongly features matter.
  • A bias is an adjustable starting push before the features are considered.
  • A loss is a wrongness score.
  • Learning means changing weights and biases to reduce loss.
  • A forward pass makes the prediction.
  • A backward pass figures out which settings should change.
  • Backpropagation is efficient blame assignment in a multi-layer network.
  • A learning rate controls how big each adjustment is.

Walkthrough

Start with features and a prediction

For the cafe order example, the inputs are:

items ordered
drink size
includes food

These inputs are features.

The model uses the features to make a prediction:

small order
large order

At first, the prediction may be bad. That is normal. Neural networks usually begin with poor settings and improve through repeated feedback.

Weights are knobs

A model can treat each feature as more or less important.

For example:

items ordered -> strong influence
drink size    -> medium influence
includes food -> strong influence

Those influence settings are weights.

If the model keeps predicting large whenever the drink is large, it may be giving the drink-size weight too much influence.

If the model ignores whether food is included, that weight may need more influence.

Bias is a starting push

The bias is another adjustable setting.

Think of it as the model's starting lean before it looks at the features.

For example:

this cafe usually has small orders

That starting lean can be useful, but it should not overpower the actual features. Training adjusts the bias too.

Loss is the wrongness score

After a prediction, the model compares its answer with the real answer.

Example:

prediction: large
real answer: small
loss: high

Another example:

prediction: almost small
real answer: small
loss: low

Loss is useful because it gives the model a signal to improve from.

The goal is not just:

be right once

The goal is:

make the loss smaller across many examples

Learning is repeated adjustment

Training repeats the same basic loop many times:

1. Look at an example.
2. Make a prediction.
3. Measure the loss.
4. Adjust the weights and bias.
5. Try again.

The model does not understand cafe orders like a person does. It improves its settings because the loss tells it whether the current settings are working.

The gradient is a direction hint

The full lesson introduces gradients.

For now, read gradient as:

which direction would make the loss go up fastest?

Since training wants lower loss, the model moves the other way.

Plain version:

gradient points uphill
training steps downhill

You do not need the derivative formula yet. The important idea is direction.

Learning rate is step size

Once the model knows a direction, it still needs to decide how far to move.

That is the learning rate.

If the learning rate is too small:

the model improves very slowly

If the learning rate is too large:

the model may jump past better settings

So learning rate is not a tiny detail. It controls whether the model adjusts steadily.

Forward pass and backward pass

Neural-network training has two directions.

The forward pass goes from inputs to prediction:

features -> hidden work -> prediction -> loss

The backward pass goes from loss back toward the earlier settings:

loss -> output settings -> hidden settings -> earlier settings

The forward pass answers:

What did the model predict?

The backward pass answers:

Which settings helped cause the error?

Backpropagation is the backward pass method

In a tiny model, it is easy to imagine adjusting one or two knobs.

Neural networks can have many layers and many weights. A later mistake may depend on many earlier settings.

Backpropagation is the efficient way to work backward through those layers and find useful correction signals.

People sometimes describe this as assigning credit or blame.

Plain version:

this output was wrong
which earlier settings contributed to that wrongness?

How backpropagation follows influence paths

The word blame is only a shortcut. Backpropagation is really following paths of influence.

Imagine the model made this path during the forward pass:

includes food -> hidden signal: probably large -> output: large

But the real answer was:

small

Now the model knows the final answer was pushed too far toward large.

Backpropagation works backward through the same path:

output was too large
-> which output settings pushed it upward?
-> which hidden signals fed those settings?
-> which input weights made those hidden signals strong?

So the logic is not random guessing.

It is closer to tracing a chain:

this value affected that value
that value affected the prediction
the prediction affected the loss

If a setting had a strong path to the mistake, it gets a larger correction. If a setting barely affected the mistake, it gets a small correction or no useful correction.

Backpropagation does not update the weights by itself. It computes the information needed for the update.

Gradient descent uses that information to change the weights.

How this connects to digit recognition

The full lesson switches to handwritten digits.

There, the features are not cafe-order facts. They are pixel values.

28 by 28 image -> 784 pixel inputs

The prediction is not small or large. It is one of ten digit classes:

0, 1, 2, 3, 4, 5, 6, 7, 8, 9

But the training loop is the same:

pixels -> prediction -> loss -> adjust weights -> repeat

That is why a simple cafe example can prepare you for the harder neural-network lesson.

Term Decoder

Use this table when reading the full lesson.

Full lesson term Read it as
feature one input fact the model can use
prediction the model's answer
weight an adjustable influence knob
bias an adjustable starting push
parameter a trainable weight or bias
loss wrongness score
gradient direction hint for how loss changes
gradient descent stepping toward lower loss
learning rate step size
forward propagation computing the prediction
backpropagation efficient blame assignment
chain rule the math rule that lets blame move backward through layers

Common traps

Do not start with the formulas

The formulas explain the mechanism later. First, understand the loop: predict, measure loss, adjust, repeat.

Do not think weights are fixed rules

Weights are learned settings. Training changes them.

Do not treat loss and accuracy as the same thing

Loss is the wrongness signal used for updates. Accuracy is a human-readable count of correct predictions.

Do not think backpropagation is the whole training process

Backpropagation computes gradient information. Training also needs predictions, loss calculation, and weight updates.

Do not treat larger steps as always better

A large learning rate can overshoot useful settings.

Check yourself

What is the basic neural-network training loop?

Make a prediction, measure the loss, adjust the weights and biases, and repeat.

In the cafe example, what are the features?

Inputs such as items ordered, drink size, and whether the order includes food.

What does a weight control?

How strongly one input influences the model's later computation.

Why does the model need a loss?

Loss tells the model how wrong the prediction was, which gives it a signal for improvement.

What is the learning rate?

The step size used when adjusting weights and biases.

What is the difference between a forward pass and a backward pass?

The forward pass computes the prediction. The backward pass works from the loss back through the network to find which settings should change.

Why is backpropagation useful in a neural network?

It efficiently assigns credit or blame through many layers and many weights.

Now Read the Full Lesson

Next: How Neural Networks Learn

When you read the full lesson, translate the main technical terms back to these ideas:

weights and biases -> adjustable knobs
loss               -> wrongness score
gradient           -> direction hint
learning rate      -> step size
forward pass       -> prediction direction
backpropagation    -> blame-assignment direction

The full lesson adds perceptron, ADALINE, gradient descent, chain rule, and the digit-recognition example.

Source anchors

  • Supports: study-guide/docs/lessons/09a-how-nns-learn.md
  • Source file: notebooks/Module2/09a-How NNs Learn.pdf
  • Key source concepts prepared here: features, weights, bias, loss, gradient, gradient descent, learning rate, forward propagation, backpropagation, chain rule, digit-recognition inputs