1 - 10 of 10 Chapters
[Einstein rides a bicycle—a concise summary of the brain’s capabilities. It can take streams of input from the vision system, the sense of balance, and the sense of place. It can combine and process this information to generate an orchestrated plan that drives myriad motor neurons to fire in a...
[“Aspects of time are often said to be more ‘abstract’ than their spatial analogues because we can perceive the spatial, but can only imagine the temporal.” ]
[At this point, we have a relatively simple model, TNNs, and the claim is that these TNNs capture the basic cognitive paradigms used in the neocortex. But, is it possible that a model for the neocortex can be this simple? If we compare the simple TNN with the incredibly complicated biological...
[TNNs as a general class are isomorphic to space-time computing systems. Considered abstractly, they exist as computing paradigms completely separate from biological neocortex operation. However, the reality is that we want to construct computing devices that can do the same kinds of things that...
[Thus far, neuron modeling has been described in general terms. In this chapter, the development of specific spiking neuron models begins. Excitatory neurons are the primary computational elements in the prototype TNN. It is these neurons that observe a volley of spikes at their inputs, process...
[Excitatory neurons perform active computation. They take spike volleys as inputs, process them, and generate spike outputs. This chapter develops a formulation for computing with excitatory neurons. Then, in later chapters this formulation (after further refinement) is implemented in the...
[In this chapter, the overall architecture of a prototype TNN is described. This architecture will be used for the extended design study in Chapter 9. The design to be undertaken in that chapter is a clustering TNN for a long-standing machine learning dataset. Although placed in separate...
[The text up to this point has focused on development of a TNN model. This chapter describes a simulator for a prototype TNN architecture. Although the simulator is intended for architecture development, the simulation code could later become the basis for an eventual machine learning...
[As a driver for developing a prototype TNN architecture, the MNIST benchmark  provides an excellent workload source. Normally, the MNIST dataset is used for classification. However, in this chapter we focus on clustering of the MNIST dataset via a TNN with unsupervised training.]
[In this book, four main assertions are put forward.]
Read and print from thousands of top scholarly journals.
Continue with Facebook
Log in with Microsoft
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Sign Up Log In
To subscribe to email alerts, please log in first, or sign up for a DeepDyve account if you don’t already have one.
To get new article updates from a journal on your personalized homepage, please log in first, or sign up for a DeepDyve account if you don’t already have one.