{"id":77,"date":"2025-08-28T00:15:14","date_gmt":"2025-08-27T22:15:14","guid":{"rendered":"https:\/\/mlatilikzsolt.hu\/?p=77"},"modified":"2025-09-06T18:44:38","modified_gmt":"2025-09-06T16:44:38","slug":"intro-to-neural-networks_part2","status":"publish","type":"post","link":"https:\/\/mlatilikzsolt.hu\/en\/2025\/08\/28\/intro-to-neural-networks_part2\/","title":{"rendered":"Introduction to the World of Neural Networks Part 2"},"content":{"rendered":"

In the previous<\/a> article, we introduced the basic idea of neural networks and saw that an artificial neuron is a simplified version of the brain\u2019s nerve cells. Now let\u2019s take a closer look at how a biological neuron works and how we can model it in a computer.<\/p>\n\n\n\n

How Does a Neuron Work?<\/h2>\n\n\n\n

The Biological Neuron<\/h3>\n\n\n\n

Without going into too much scientific detail, a biological neuron is made up of four main parts:<\/p>\n\n\n\n

    \n
  • Dendrites: it receives information from other neurons through these.<\/li>\n\n\n\n
  • Cell body (Soma): this processes the signals received by the dendrites.<\/li>\n\n\n\n
  • Axon: the neuron sends (or not) the processed signal through this.<\/li>\n\n\n\n
  • Axon terminals: the branches of the axon through which other neurons perceive the output signal.<\/li>\n<\/ul>\n\n\n\n
    \"\"<\/figure>\n\n\n\n

    So, the cell body receives signals from other neurons through the dendrites, processes them, and if the signal resulting from the processing reaches a certain level, the neuron \"fires\", i.e. sends a signal through the axon to the other neurons connected to it.<\/p>\n\n\n\n

    Artificial Neuron<\/h3>\n\n\n\n

    The artificial neuron attempts to mimic this operation mathematically. Its most important parts are:<\/p>\n\n\n\n

      \n
    • Inputs: these simulate the dendrites, through which the neuron receives data.<\/li>\n\n\n\n
    • Weights: each input has a weight that shows how much the data arriving at the given input influences the output value.<\/li>\n\n\n\n
    • Bias: an offset is added to the weighted sum of the input signals, which can also influence the output result.<\/li>\n\n\n\n
    • Activation function: this makes the decision as to what value should be output by the neuron (\"fire\" or not) based on the previously summarized data.<\/li>\n<\/ul>\n\n\n\n
      \"\"<\/figure>\n\n\n\n

      In Mathematical Form<\/h4>\n\n\n\n

      Let the inputs be in order x1<\/sub>, x2<\/sub>…xn<\/sub><\/strong>, their corresponding weights w1<\/sub>, w2<\/sub>…wn<\/sub><\/strong>, and the bias b<\/strong>. The operation performed by the neuron is as follows:<\/p>\n\n\n\n

      z=w_1 \\cdot x_1 + w_2 \\cdot x_2 + \\ldots + w_n \\cdot x_n + b<\/pre><\/div>\n\n\n\n
      The activation function receives the value of z<\/strong> calculated in this way. In this case, let's take a simple step<\/em> function that examines the input value and if it is zero or greater, it outputs 1, and if it is less than zero, it outputs 0.<\/p>\n\n\n\n
      y=\\begin{cases} 1 & \\text{ha } z \\geq 0 \\\\ 0 & \\text{ha } z < 0 \\end{cases}<\/pre><\/div>\n\n\n\n
      There are many types of activation functions (e.g. Sigmoid, ReLU, tanh<\/em>), which we will discuss in a separate section later.<\/p>\n\n\n\n
      Python Example<\/h2>\n\n\n\n
      Let's see how to program a neuron in Python:<\/p>\n\n\n\n