hAIgopala › AI terms › Rapid Fire (3 terms per short) › Day 10

Activation Function (ReLU, Sigmoid, Softmax)

Neural Networks & Deep LearningChapter 3: Inside the BrainTerm 43 of 253

Adds non-linearity so networks can learn complex patterns. Softmax turns scores into probabilities.

At Sunrise School

Gopal asks

How does a neuron decide to fire?

Bit answers

An activation function. ReLU keeps positive signals and drops negatives; softmax turns final scores into probabilities like Fees 87%.

In the real world

ReLU: if negative → 0, else keep.

Watch the short

Short coming soonSeason 5 · Day 10Follow @haigopala

Quick check

Quick check. Is a bigger learning rate always faster to a good model? Yes or no?

Taught together with

Learning Rate · Epoch

Practise it

Try this term and all 253 others in the free Model Training Arena.

Open the Arena →

Follow Bit daily: YouTube English · తెలుగు · हिंदी  |  Instagram English · తెలుగు · हिंदी