An associative memory retrieves stored information by its content rather than by an address: a partial or noisy cue is enough to recall the matching whole – much as the opening bars bring back an entire song. Hopfield networks are a classic example in AI research. The principle also appears in hardware (content-addressable memory) and in analyses of the attention mechanism in modern language models.
Artificial neural networks (ANNs) are computational models made up of many connected units (“neurons”) arranged in layers that process inputs via weighted connections. During training, the weights are adjusted step by step using examples, usually by backpropagation. ANNs are the foundation of deep learning and power image recognition, speech recognition and large language models. Despite their biological inspiration, they work far more simply than the human brain.
Pattern recognition is the automatic identification of regularities, structures or recurring features in data such as images, sounds, texts or measurements. Systems assign new observations to known categories or discover groups without predefined labels. Applications include facial recognition, handwriting recognition, quality inspection in manufacturing and spotting unusual transactions. The discipline is an important foundation of machine learning.
Cognitive computing is an umbrella term for systems intended to emulate human thought processes such as understanding language, reasoning and learning from experience in order to support people's decisions. The term became widely known in the 2010s, mainly through IBM Watson. It combines methods such as natural language processing, machine learning and knowledge representation; today people more often speak of AI assistants or language models instead.
