Exercise 1 – Modern Artificial Neuron

The modern artificial neuron is the basic building block of current neural networks such as MLPs, CNNs, and transformers. Which of the following best describes a modern artificial neuron used in today's neural networks?

Drag these lines into the blanks:

  • It uses binary inputs and a fixed step function like McCulloch-Pitts.
  • It multiplies inputs by weights, adds bias, and passes the result through a nonlinear activation function.
  • It works without any weights or bias and returns a constant value.
  • It requires training with only logical rules, not gradient-based methods.
Drop here
Scroll to Top