Modelling Emergent Phenomena In Associative Memory With Neural Cell Assemblies 1st Kailash Nadh by Kailash Nadh instant download after payment.

Dissertation for the pursuit of the Doctor of Philosophy degree 

This thesis is an exploratory study of the dynamics of CAs and CA based associative memory. It looks into how complex phenomena emerge from associative memories by modelling and simulating the cognitive processes of formation, association, and retrieval of memories as CAs. From these elemental processes, higher order behaviour are obtained, namely, emergent context sensitivity, spatial cognitive mapping (modelled in an embodied virtual agent), and emergence of novel behaviour (explored via an autonomous game playing agent). 

The thesis also confirms that CAs are capable of performing real world tasks, with a natural language processing model capable of resolving with better accuracy than many symbolic machine learning models, the prepositional phrase attachment ambiguity, a common syntactic and semantic ambiguity.

The results from the models are novel and manifold, hinting at a unified model of associative memory. They suggest that neurobiologically inspired models—in particular, the CA model—may be better at performing certain AI tasks than other traditional computational models. They also suggest the possibility of a CA based associative memory model that may be able to account for many higher order processes, and demonstrate how CAs can be used to model tasks in AI that resemble processes in the brain."