Abstract
The Semantic Pointer Architecture (SPA) is a proposal of specifying the computations and architectural elements needed to account for cognitive functions. By means of the Neural Engineering Framework (NEF) this proposal can be realized in a spiking neural network. However, in any such network each SPA transformation will accumulate noise. By increasing the accuracy of common SPA operations, the overall network performance can be increased considerably. As well, the representations in such networks present a trade-off between being able to represent all possible values and being only able to represent the most likely values, but with high accuracy. We derive a heuristic to find the near-optimal point in this trade-off. This allows us to improve the accuracy of common SPA operations by up to 25 times. Ultimately, it allows for a reduction of neuron number and a more efficient use of both traditional and neuromorphic hardware, which we demonstrate here.
Highlights
The Neural Engineering Framework (NEF) [1] is a mathematical theory of how biological neural systems can implement a wide variety of dynamic functions
This proposal is called the Semantic Pointer Architecture (SPA), and suggests specific computations, architectural elements, and methods of representing and transmitting information to account for perceptual, motor, and cognitive behaviour
The methods in this paper are applicable to the Neural Engineering Framework (NEF) [1] which allows the construction of large-scale neural networks from a mathematical description
Summary
The Neural Engineering Framework (NEF) [1] is a mathematical theory of how biological neural systems can implement a wide variety of dynamic functions. These methods have been used to propose novel models of a wider variety of neural systems, including the barn owl auditory system [2, 3], parts of the rodent navigation system [4], escape and swimming control in zebrafish [5], tactile working memory in monkeys [6], and simple decision making in humans [7] and rats [8]. The NEF provides a method for capturing how neural computations might be performed. The SPA was used to construct what remains the largest functional brain model, called Spaun [10]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.