Neurons, Weights, and Layers

Summary

Understand how artificial neurons, weights, and layers form the building blocks of neural networks.

basic
neural-network-basics

Neural networks are made of artificial neurons, which mimic how the brain processes information.

Each neuron takes inputs, multiplies them by weights, and passes them forward.

🔩 How a Neuron Works

A simple neuron performs this operation:

output = activation(w₁x₁ + w₂x₂ + ... + b)

Where:

  • x = inputs
  • w = weights
  • b = bias
  • activation() = function that adds non-linearity

🔁 Weights and Bias

  • Weights determine how much influence each input has.
  • Bias allows shifting the output up or down.

Training adjusts these values to reduce loss.

🧱 Layers of Neurons

Neural networks are organized in layers:

  1. Input Layer — receives the raw data
  2. Hidden Layers — learn patterns and features
  3. Output Layer — produces final predictions

📊 Example Network

  • Each circle = neuron
  • Each line = weight

🧠 Why Layers Matter

  • More layers → More expressive power
  • Too many → Risk of overfitting
  • Finding the balance is key

🧠 Summary

| Concept | Role | |----------|--------------------------------------| | Neuron | Basic computing unit | | Weight | Multiplier that adjusts input signal | | Bias | Shifts the output | | Layer | Group of neurons |

✅ Self-Check

  • What does a single neuron compute?
  • What are weights and why are they important?
  • What is the role of hidden layers?