Limits...
Multiclass Classification by Adaptive Network of Dendritic Neurons with Binary Synapses Using Structural Plasticity.

Hussain S, Basu A - Front Neurosci (2016)

Bottom Line: The performance of the model is evaluated on classification of handwritten digits from the benchmark MNIST dataset and compared with other spike classifiers.For VLSI implementations, we show that the reduced synaptic memory can save upto 4X area compared to conventional crossbar topologies.Finally, we also present a biologically realistic spike-based version for calculating the correlations required by the structural learning rule and demonstrate the correspondence between the rate-based and spike-based methods of learning.

View Article: PubMed Central - PubMed

Affiliation: School of Electrical and Electronic Engineering, Nanyang Technological University Singapore, Singapore.

ABSTRACT
The development of power-efficient neuromorphic devices presents the challenge of designing spike pattern classification algorithms which can be implemented on low-precision hardware and can also achieve state-of-the-art performance. In our pursuit of meeting this challenge, we present a pattern classification model which uses a sparse connection matrix and exploits the mechanism of nonlinear dendritic processing to achieve high classification accuracy. A rate-based structural learning rule for multiclass classification is proposed which modifies a connectivity matrix of binary synaptic connections by choosing the best "k" out of "d" inputs to make connections on every dendritic branch (k < < d). Because learning only modifies connectivity, the model is well suited for implementation in neuromorphic systems using address-event representation (AER). We develop an ensemble method which combines several dendritic classifiers to achieve enhanced generalization over individual classifiers. We have two major findings: (1) Our results demonstrate that an ensemble created with classifiers comprising moderate number of dendrites performs better than both ensembles of perceptrons and of complex dendritic trees. (2) In order to determine the moderate number of dendrites required for a specific classification problem, a two-step solution is proposed. First, an adaptive approach is proposed which scales the relative size of the dendritic trees of neurons for each class. It works by progressively adding dendrites with fixed number of synapses to the network, thereby allocating synaptic resources as per the complexity of the given problem. As a second step, theoretical capacity calculations are used to convert each neuronal dendritic tree to its optimal topology where dendrites of each class are assigned different number of synapses. The performance of the model is evaluated on classification of handwritten digits from the benchmark MNIST dataset and compared with other spike classifiers. We show that our system can achieve classification accuracy within 1 - 2% of other reported spike-based classifiers while using much less synaptic resources (only 7%) compared to that used by other methods. Further, an ensemble classifier created with adaptively learned sizes can attain accuracy of 96.4% which is at par with the best reported performance of spike-based classifiers. Moreover, the proposed method achieves this by using about 20% of the synapses used by other spike algorithms. We also present results of applying our algorithm to classify the MNIST-DVS dataset collected from a real spike-based image sensor and show results comparable to the best reported ones (88.1% accuracy). For VLSI implementations, we show that the reduced synaptic memory can save upto 4X area compared to conventional crossbar topologies. Finally, we also present a biologically realistic spike-based version for calculating the correlations required by the structural learning rule and demonstrate the correspondence between the rate-based and spike-based methods of learning.

No MeSH data available.


Dendritic classifier is trained adaptively by adding dendrites (step-1) and keeping the number of synapses/dendrite (k) as a constant for all neurons. The total number of synapses for the μth neuron learned in this manner () is fixed and the corresponding optimal topology (,  inset plot) is theoretically determined (step-2). The length of sides of the rectangle denote m and k values and the area represents the total number of synapses, s.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4814530&req=5

Figure 3: Dendritic classifier is trained adaptively by adding dendrites (step-1) and keeping the number of synapses/dendrite (k) as a constant for all neurons. The total number of synapses for the μth neuron learned in this manner () is fixed and the corresponding optimal topology (, inset plot) is theoretically determined (step-2). The length of sides of the rectangle denote m and k values and the area represents the total number of synapses, s.

Mentions: This adaptive learning method is guided by the level of difficulty of each neuron's classification task and moreover, it is used to learn the relative size of each neuron and not the optimal neuron topology. While performing this adaptation, we keep the number of synapses per branch, k, to be a constant for all neurons and just vary the number of dendrites per neuron, m. Hence, at the end of the adaptation, both PDT and NDT of the μth class have a total number of synapses given by , where is the number of dendritic branches learned for the μth class neurons after adaptation. This is shown as the step-1 in Figure 3, where the topology of a class neuron is denoted by the rectangle with its sides representing m and k values.


Multiclass Classification by Adaptive Network of Dendritic Neurons with Binary Synapses Using Structural Plasticity.

Hussain S, Basu A - Front Neurosci (2016)

Dendritic classifier is trained adaptively by adding dendrites (step-1) and keeping the number of synapses/dendrite (k) as a constant for all neurons. The total number of synapses for the μth neuron learned in this manner () is fixed and the corresponding optimal topology (,  inset plot) is theoretically determined (step-2). The length of sides of the rectangle denote m and k values and the area represents the total number of synapses, s.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4814530&req=5

Figure 3: Dendritic classifier is trained adaptively by adding dendrites (step-1) and keeping the number of synapses/dendrite (k) as a constant for all neurons. The total number of synapses for the μth neuron learned in this manner () is fixed and the corresponding optimal topology (, inset plot) is theoretically determined (step-2). The length of sides of the rectangle denote m and k values and the area represents the total number of synapses, s.
Mentions: This adaptive learning method is guided by the level of difficulty of each neuron's classification task and moreover, it is used to learn the relative size of each neuron and not the optimal neuron topology. While performing this adaptation, we keep the number of synapses per branch, k, to be a constant for all neurons and just vary the number of dendrites per neuron, m. Hence, at the end of the adaptation, both PDT and NDT of the μth class have a total number of synapses given by , where is the number of dendritic branches learned for the μth class neurons after adaptation. This is shown as the step-1 in Figure 3, where the topology of a class neuron is denoted by the rectangle with its sides representing m and k values.

Bottom Line: The performance of the model is evaluated on classification of handwritten digits from the benchmark MNIST dataset and compared with other spike classifiers.For VLSI implementations, we show that the reduced synaptic memory can save upto 4X area compared to conventional crossbar topologies.Finally, we also present a biologically realistic spike-based version for calculating the correlations required by the structural learning rule and demonstrate the correspondence between the rate-based and spike-based methods of learning.

View Article: PubMed Central - PubMed

Affiliation: School of Electrical and Electronic Engineering, Nanyang Technological University Singapore, Singapore.

ABSTRACT
The development of power-efficient neuromorphic devices presents the challenge of designing spike pattern classification algorithms which can be implemented on low-precision hardware and can also achieve state-of-the-art performance. In our pursuit of meeting this challenge, we present a pattern classification model which uses a sparse connection matrix and exploits the mechanism of nonlinear dendritic processing to achieve high classification accuracy. A rate-based structural learning rule for multiclass classification is proposed which modifies a connectivity matrix of binary synaptic connections by choosing the best "k" out of "d" inputs to make connections on every dendritic branch (k < < d). Because learning only modifies connectivity, the model is well suited for implementation in neuromorphic systems using address-event representation (AER). We develop an ensemble method which combines several dendritic classifiers to achieve enhanced generalization over individual classifiers. We have two major findings: (1) Our results demonstrate that an ensemble created with classifiers comprising moderate number of dendrites performs better than both ensembles of perceptrons and of complex dendritic trees. (2) In order to determine the moderate number of dendrites required for a specific classification problem, a two-step solution is proposed. First, an adaptive approach is proposed which scales the relative size of the dendritic trees of neurons for each class. It works by progressively adding dendrites with fixed number of synapses to the network, thereby allocating synaptic resources as per the complexity of the given problem. As a second step, theoretical capacity calculations are used to convert each neuronal dendritic tree to its optimal topology where dendrites of each class are assigned different number of synapses. The performance of the model is evaluated on classification of handwritten digits from the benchmark MNIST dataset and compared with other spike classifiers. We show that our system can achieve classification accuracy within 1 - 2% of other reported spike-based classifiers while using much less synaptic resources (only 7%) compared to that used by other methods. Further, an ensemble classifier created with adaptively learned sizes can attain accuracy of 96.4% which is at par with the best reported performance of spike-based classifiers. Moreover, the proposed method achieves this by using about 20% of the synapses used by other spike algorithms. We also present results of applying our algorithm to classify the MNIST-DVS dataset collected from a real spike-based image sensor and show results comparable to the best reported ones (88.1% accuracy). For VLSI implementations, we show that the reduced synaptic memory can save upto 4X area compared to conventional crossbar topologies. Finally, we also present a biologically realistic spike-based version for calculating the correlations required by the structural learning rule and demonstrate the correspondence between the rate-based and spike-based methods of learning.

No MeSH data available.