Introduction to Deep-Learning Concepts and TensorFlow
Santanu Pattanayak · Apress eBooks · 2022
Deep learning evolved from artificial neural networks, which have existed since the 1940s. Neural networks are interconnected networks of processing units called artificial neurons that loosely mimic the axons found in a biological brain. In a biological neuron, dendrites receive input signals from various neighboring neurons, typically more than 1000 of them. These modified signals are then passed on to the cell body or soma of the neuron, where these signals are summed together and then passed on to the axon of the neuron. If the received input signal is more than a specified threshold, the axon will release a signal, which will be passed on to the neighboring dendrites of other neurons. Figure 2-1 depicts the structure of a biological neuron for reference.