A Proposal And Implementation Of A Neural Network Based Hierarchical Temporal Memory To Realize Cognitive Functions

Hierarchical Temporal Memory (HTM) is a recent innovation in cognition science. Developedrnin 2005 by Numenta Inc., an artificial intelligence research firm in the US, HTMs attempt torncapture the way the human brain learns and infers its environment. One of the most notablerncharacteristics of this model is the consideration of the hierarchical organization of objects inrnthe world. Data in the world is made up of elementary features that aggregate in successivernlayers to form perceivable objects. This data can be visual, auditory or from other abstractrnspaces such as stock markets and scientific studies. The amount of raw data that the brain isrnexposed to throughout its lifetime is beyond imagination. However the brain is known to use arnvery noble and systematic approach to handle the perception, storage, and inference of thisrndata. Several studies in neuroscience and psychology indicate that the brain makes use of thernhierarchies that features in the world exhibit in their organization to form objects. Hence, forrninstance, ‘corners’ and ‘lines’ can aggregate to form a ‘table’ object in the visual world. Thesernelementary features, however, can use a different aggregation to form a ‘chair’ object. Thernsame is true for data in other types of worlds such as audio. HTMs directly apply a similarrnhandling of world data for their cognition. Furthermore, the structure of HTMs, made up ofrndata processing nodes arranged in a hierarchical tree, mimic the physical arrangement ofrncortical layers in the brain.rnThe various data analysis algorithms in the nodes of HTMs were, however, found by thernresearcher to be limiting in several aspects, one of which is handling of unforeseen (untrained)rndata. For this purpose neural network data structures were used to replace some of thernoperations in these nodes for their ability in approximating untrained data to nearest matches.rnThis research work discusses the proposed model, implementation constraints, the operationalrncharacteristics, and performance enhancements observed in this modified model with arnselected test application. A substantial improvement in cognitive functionality has beenrnobserved with the newly proposed model.

Get Full Work

Report copyright infringement or plagiarism