I like how Pedro Domingos, in his The Master Algorithm book, describes how in the AI field you have to first approximation these camps: - Connectionists like to mimic the brain's interconnected neurons, mostly simpler models (neuroscience): artificial neural networks, deep learning, spiking neural networks, liquid neural networks, neuromorphic computing,... (Deep learning enjoyers, Sandamiskaya) - Symbolists like symbol manipulation: decision trees, random decision forests, production rule systems, inductive logic programming,... - Bayesians like uncertainity reduction based on probability theory (staticians): bayes classifier, probabilistic graphical models, hidden markov chains, active inference,... (Jeff Beck, Friston) Frequentists exist too with their own definition of probability. - Evolutionaries like evolution (biologists): genetic algorithms, evolutionary programming (Stanley) - Analogizers like identifying similarities between situations or things (psychologists): k-nearest neighbors, support vector machines,... Then there are various hybrids: neurosymbolic architectures (AlphaZero for chess, general program synthesis with DreamCoder) (Chollet, Gary Marcus), neuroevolution, bayesian deep learning, etc. And I would add that technically you can also have: - Reinforcement Learners like learning from reinforcement signals: reinforcement learning (AlphaZero, LLM RLVR,...) (Sutton) - Causal Inferencers like to build a causal model and can thereby make inferences using causality rather than just correlation: causal AI (Pearl) - Compressionists who see cognition as a form of compression: autoencoders, huffman encoding, Hutter prize (Hutter) - Divergent Objectiveless/Novelty Searchers love divergent search for novelty without objectives: novelty search (Stanley) - Brainers camp that likes to figure out and implement (mimic) both the architecture and the learning algorithms of the brain more closely (NeuroAI subfield, neuromorphic computing, spiking neural networks, liquid neural networks, predictive coding, hebbian learning, forward forward algorithm, simulating hippocampus, hodgkin-huxley model,...) - physicists liking physics concepts (diffusion models, flow matching, Schrodinger bridges, "there will be science of deep learning" learning mechanics people, geometric deep learning) - mathematicians liking applied/pure mathy concepts (categorical deep learning, geometric deep learning). - "Theory of Computationaries" camp that likes architectures like neural turing machines - Scalers, where subtribes and subsubtribes of this tribe disagree on what exactly to scale (more or less symbolic aspects in neurosymbolic system, more or less brain inspired aspects, what specific deep learning architecture,...) And you can hybridize these too with deep reinforcement learning, novelty search with other objectives, causal deep learning, etc. A lot of those labels of camps are often approximate fuzzy high level clusters, often with no clear boundaries, and various degrees of overlap If you're curious about the details of these architectures, you can search them easily on Google search, or ask your favorite autoregressive transformer LLM. [The Master Algorithm - Wikipedia](https://en.wikipedia.org/wiki/The_Master_Algorithm)