Artificial neural network
Artificial neural network is a computational structure modeled after the neuron connections in the brain. This is the basic material for deep learning.
Artificial neural network is a computational structure inspired by the way neurons exchange signals in the brain. Numerous artificial neurons are connected layer by layer, and the strength (weight) of each connection is learned by gradually adjusting it according to the data. It can be likened to the process of adjusting the numerous dials on a radio until the sound is the clearest.
It was created to learn from data on problems that are difficult to program rules individually, such as distinguishing between a cat and a dog in a photo. Deep learning is deep learning, and it is the technology that forms the basis of modern AI, from image recognition to ChatGPT.
It is said to be modeled after the brain, but a common misunderstanding is that it is quite different from how the actual brain works. Although it is mathematically inspired, an artificial neural network is not an artificial brain.
✅ Why it matters
- Learn by yourself from data to solve problems that are difficult to solve with rules
- It is a common foundation for almost all AI fields, including images, voice, and language
- It is a starting point for understanding news related to deep learning and LLM
⚠️ Limits and debates
- It has a black box nature that makes it difficult for humans to interpret the internal judgment basis
- A lot of data and computational resources are needed to achieve good performance
- The expression that it imitates the brain is often taken as exaggerated than it actually is.