1. Hierarchical Temporal Memory (HTM): HTM is a biologically-inspired computational model of cortical circuits developed by Jeff Hawkins and his team at Numenta. It is designed to learn and recognize patterns in sensory input and to perform predictive modeling.
  2. The Bayesian Brain Theory: The Bayesian Brain Theory proposes that the brain represents the world probabilistically and uses Bayesian inference to make predictions and decisions. It is based on the idea that the brain constructs a probabilistic model of the world based on sensory input and prior knowledge.
  3. The Predictive Coding Theory of Cognition: The Predictive Coding Theory of Cognition proposes that the brain is a hierarchical generative model that uses top-down predictions to interpret sensory input. It suggests that perception and action are intertwined and that the brain actively constructs a model of the world based on prior experience and current sensory input.
  4. Evolutionary Computation: Evolutionary computation is a class of optimization algorithms inspired by biological evolution. It involves creating populations of candidate solutions and selecting the best solutions for the next generation based on their fitness. This approach has been used to develop artificial neural networks and other machine learning models, and has been proposed as a possible route to AGI.
  5. The Neural Engineering Framework: The Neural Engineering Framework is a theoretical model developed by Chris Eliasmith and his team at the University of Waterloo. It involves designing neural models that can perform complex cognitive tasks by integrating sensory information, making decisions, and generating motor output. The approach is based on the idea that the brain is a complex dynamical system that can be modeled using principles of systems engineering.
  1. AIXI by Hutter: AIXI is a theoretical model for AGI based on Solomonoff induction, which is a formal theory of inductive inference. AIXI is an agent that takes actions in an environment to maximize its reward. It uses a probability distribution over possible environments and policies to make decisions about what actions to take.
  2. Neural Turing Machines (NTMs) by Graves et al.: NTMs are a type of neural network that can read and write to an external memory. They are inspired by Turing machines, which are theoretical models of computation that can perform any algorithm. NTMs are capable of learning algorithms from examples and can perform tasks that require complex reasoning and manipulation of symbols.
  3. Universal AI (UAI) by Goertzel and Pitts: UAI is a theoretical model for AGI that combines several different approaches to AI, including logic-based reasoning, probabilistic reasoning, and neural networks. UAI is designed to be a universal learning agent that can adapt to different environments and tasks.
  4. The Recursive Cortical Network (RCN) by Tononi and Edelman: The RCN is a computational model of the brain that is based on the theory of neuronal group selection. It consists of a network of neurons that are organized into functional groups. The RCN is capable of learning and adapting to new tasks and environments.
  5. The Sparse Distributed Memory (SDM) by Kanerva: The SDM is a computational model of memory that is based on the idea that memories are represented as sparse and distributed patterns of activity in the brain. The SDM can store and retrieve large amounts of information and is capable of generalizing from past experiences.
  6. The Global Workspace Theory (GWT) by Baars: The GWT is a theoretical model of consciousness that proposes that consciousness arises from the integration of information across multiple brain regions. The GWT suggests that an AGI would need a similar mechanism for integrating information from different sources.
  7. The Society of Mind (SoM) by Minsky: The SoM is a theoretical model of intelligence that proposes that intelligence is not a single, unified entity, but rather a collection of smaller, specialized agents that work together to solve problems. An AGI based on the SoM would be composed of many different modules that work together to achieve a common goal.
  8. The Bayesian Brain Theory (BBT) by Friston: The BBT is a theoretical model of the brain that proposes that the brain is constantly updating its beliefs about the world based on incoming sensory information. The BBT suggests that an AGI would need a similar mechanism for updating its beliefs about the world.
  9. The Integrated Information Theory (IIT) by Tononi: The IIT is a theoretical model of consciousness that proposes that consciousness arises from the integration of information across a complex network of brain regions. The IIT suggests that an AGI would need a similar mechanism for integrating information across multiple domains.
  10. The Dynamic Core Hypothesis (DCH) by Edelman: The DCH is a theoretical model of cognition that proposes that cognition arises from the dynamic interactions between multiple brain regions. The DCH suggests that an AGI would need a similar mechanism for coordinating its cognitive processes.
  11. The Hierarchical Temporal Memory (HTM) by Hawkins: The HTM is a theoretical model of the neocortex that is based on the idea that the neocortex performs hierarchical pattern recognition. The HTM suggests that an AGI would need a similar mechanism for hierarchical pattern recognition.
  12. The Complementary Learning Systems (CLS) theory by McClelland and Rogers: The CLS theory is a theoretical model of learning that proposes that the brain has two complementary learning systems, one that is fast and one that is slow. The CLS theory suggests that an AGI would need a similar mechanism for learning.

Each of these models has its own unique perspective on how AGI might be achieved, and researchers continue to explore new approaches to developing AGI. Peraphs AGI will be achieved by one of models, or a mixture of methods and concepts, or perhaps - a completely new framework that is equally paradigm shifting of the ones i’ve stated above