Artificial neural networks comprising spiking neurons of a novel type have been conceived as improved pattern- analysis and pattern- recognition computational systems. These neurons are represented by a mathematical model denoted the state- variable model (SVM), which among other things, exploits a computational parallelism inherent in spiking-neuron geometry. Networks of SVM neurons offer advantages of speed and computational efficiency, relative to traditional artificial neural networks. The SVM also overcomes some of the limitations of prior spiking-neuron models. There are numerous potential pattern-recognition, tracking, and data-reduction (data preprocessing) applications for these SVM neural networks on Earth and in exploration of remote planets.
Spiking neurons imitate biological neurons more closely than do the neurons of traditional artificial neural networks. A spiking neuron includes a central cell body (soma) surrounded by a treelike interconnection network (dendrites). Spiking neurons are so named because they generate trains of output pulses (spikes) in response to inputs received from sensors or from other neurons. They gain their speed advantage over traditional neural networks by using the timing of individual spikes for computation, whereas traditional artificial neurons use averages of activity levels over time. Moreover, spiking neurons use the delays inherent in dendritic processing in order to efficiently encode the information content of incoming signals. Because traditional artificial neurons fail to capture this encoding, they have less processing capability, and so it is necessary to use more gates when implementing traditional artificial neurons in electronic circuitry. Such higher-order functions as dynamic tasking are effected by use of pools (collections) of spiking neurons interconnected by spike-transmitting fibers.
The SVM includes adaptive thresholds and submodels of transport of ions (in imitation of such transport in biological neurons). These features enable the neurons to adapt their responses to high-rate inputs from sensors, and to adapt their firing thresholds to mitigate noise or effects of potential sensor failure. The mathematical derivation of the SVM starts from a prior model, known in the art as the point soma model, which captures all of the salient properties of neuronal response while keeping the computational cost low. The point-soma latency time is modified to be an exponentially decaying function of the strength of the applied potential.
Choosing computational efficiency over biological fidelity, the dendrites surrounding a neuron are represented by simplified compartmental submodels and there are no dendritic spines. Updates to the dendritic potential, calcium- ion concentrations and conductances, and potassium-ion conductances are done by use of equations similar to those of the point soma. Diffusion processes in dendrites are modeled by averaging among nearest-neighbor compartments. Inputs to each of the dendritic compartments come from sensors.Alternatively or in addition, when an affected neuron is part of a pool, inputs can come from other spiking neurons.
At present, SVM neural networks are implemented by computational simulation, using algorithms that encode the SVM and its submodels. However, it should be possible to implement these neural networks in hardware: The differential equations for the dendritic and cellular processes in the SVM model of spiking neurons map to equivalent circuits that can be implemented directly in analog very-large-scale integrated (VLSI) circuits.
This work was done by Terrance Huntsberger of Caltech for NASA's Jet Propulsion Laboratory.
NPO- 40945
This Brief includes a Technical Support Package (TSP).
Spiking Neurons for Analysis of Patterns
(reference NPO-40945) is currently available for download from the TSP library.
Don't have an account?
Overview
The document titled "Spiking Neurons for Analysis of Patterns" from NASA's Jet Propulsion Laboratory discusses the advancements in spiking neuron technology and its applications in pattern recognition and data processing. Spiking neurons, inspired by biological neural systems, utilize the timing of spikes to represent and process information more efficiently than traditional neural networks.
Key highlights include the performance advantages of spiking neuron algorithms over conventional methods. Preliminary studies indicate that spiking neuron models can outperform traditional correlation matching approaches by an order of magnitude in pattern recognition tasks, achieving response times of 120 milliseconds compared to 1 second for conventional methods. Additionally, spiking neurons demonstrate resilience to data loss, maintaining performance even with a 5% reduction in sensor data.
The document emphasizes the potential of spiking neurons in various applications, particularly in space missions where efficient data processing is crucial. For instance, the Ground Penetrating Radar (GPR) instrument on the JIMO (Jupiter Icy Moons Orbiter) mission is expected to generate terabytes of data. Spiking neurons offer innovative data representation methods that significantly enhance storage capacity and processing efficiency. These methods include count code, timing code, rank order code, and synchrony code, which can represent vastly more items than traditional binary systems.
The spiking neuron technology is particularly beneficial for dynamic retasking in real-time scenarios, such as on-the-fly adjustments in response to changing conditions during missions. The document outlines how coupled pools of neurons can reduce turnaround times for planning from 60-600 seconds to just 10-30 seconds, enabling more agile responses to scientific targets.
Furthermore, the document references various studies that have explored the use of spiking neurons in image processing and segmentation, highlighting their effectiveness in handling complex visual data. The leaky integrate-and-fire (LIF) model, a simplified version of the Hodgkin-Huxley model, serves as the foundation for many spiking neuron studies, although it has limitations in sustaining activity.
In conclusion, the document presents spiking neurons as a transformative technology for pattern recognition and data processing, with significant implications for future space missions and beyond. The advancements in this field promise to enhance the capabilities of onboard systems, enabling more efficient and effective scientific exploration.
