Backpropagation was invented in the 1970s as a general optimization method for automatically differentiating complex nested functions. However, it wasn't until 1986, when David Rumelhart, Geoffrey Hinton, and Ronald Williams published a paper in the journal Nature titled "Learning Representations by Back-Propagating Errors," that the importance of the algorithm was appreciated. The paper introduced its algorithm and explained its vital role in deep learning. This machine learning paradigm has revolutionized various fields, including computer vision, natural language processing, and speech recognition. The backpropagation algorithm computes the network's output error and then propagates this error backward through the layers. Backpropagation's popularity has recently surged, given the widespread adoption of deep neural networks for image and speech recognition.