'Is neural network training done with one sample at a time or with an average of multiple samples?

I recently saw this video (https://youtu.be/Ilg3gGewQ5U), where the guy says that the weights in a neural network are adjusted based on an AVERAGE of the errors of multiple samples (at 8:37, if you don't want to see the whole video). Is this always the case, or are there situations where it is better to update the weights after every sample?

c++sharpdeep-learningneural-network


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions