Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell
The perceptron learning algorithm and its multiple-layer extension, the backpropagation algorithm, are the foundations of the present-day machine learning revolution. However, these algorithms utilize a highly simplified mathematical abstraction of a neuron; it is not clear to what extent real bioph...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-04-01
|
| Series: | Frontiers in Computational Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fncom.2020.00033/full |
| _version_ | 1828765225062498304 |
|---|---|
| author | Toviah Moldwin Idan Segev Idan Segev |
| author_facet | Toviah Moldwin Idan Segev Idan Segev |
| author_sort | Toviah Moldwin |
| collection | DOAJ |
| description | The perceptron learning algorithm and its multiple-layer extension, the backpropagation algorithm, are the foundations of the present-day machine learning revolution. However, these algorithms utilize a highly simplified mathematical abstraction of a neuron; it is not clear to what extent real biophysical neurons with morphologically-extended non-linear dendritic trees and conductance-based synapses can realize perceptron-like learning. Here we implemented the perceptron learning algorithm in a realistic biophysical model of a layer 5 cortical pyramidal cell with a full complement of non-linear dendritic channels. We tested this biophysical perceptron (BP) on a classification task, where it needed to correctly binarily classify 100, 1,000, or 2,000 patterns, and a generalization task, where it was required to discriminate between two “noisy” patterns. We show that the BP performs these tasks with an accuracy comparable to that of the original perceptron, though the classification capacity of the apical tuft is somewhat limited. We concluded that cortical pyramidal neurons can act as powerful classification devices. |
| first_indexed | 2024-12-11T06:44:22Z |
| format | Article |
| id | doaj.art-64239bc2d22b4644ac8ba93d8e44032c |
| institution | Directory Open Access Journal |
| issn | 1662-5188 |
| language | English |
| last_indexed | 2024-12-11T06:44:22Z |
| publishDate | 2020-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computational Neuroscience |
| spelling | doaj.art-64239bc2d22b4644ac8ba93d8e44032c2022-12-22T01:17:07ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882020-04-011410.3389/fncom.2020.00033500178Perceptron Learning and Classification in a Modeled Cortical Pyramidal CellToviah Moldwin0Idan Segev1Idan Segev2Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem, Jerusalem, IsraelEdmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem, Jerusalem, IsraelDepartment of Neurobiology, The Hebrew University of Jerusalem, Jerusalem, IsraelThe perceptron learning algorithm and its multiple-layer extension, the backpropagation algorithm, are the foundations of the present-day machine learning revolution. However, these algorithms utilize a highly simplified mathematical abstraction of a neuron; it is not clear to what extent real biophysical neurons with morphologically-extended non-linear dendritic trees and conductance-based synapses can realize perceptron-like learning. Here we implemented the perceptron learning algorithm in a realistic biophysical model of a layer 5 cortical pyramidal cell with a full complement of non-linear dendritic channels. We tested this biophysical perceptron (BP) on a classification task, where it needed to correctly binarily classify 100, 1,000, or 2,000 patterns, and a generalization task, where it was required to discriminate between two “noisy” patterns. We show that the BP performs these tasks with an accuracy comparable to that of the original perceptron, though the classification capacity of the apical tuft is somewhat limited. We concluded that cortical pyramidal neurons can act as powerful classification devices.https://www.frontiersin.org/article/10.3389/fncom.2020.00033/fullcompartmental modelingnon-linear dendritescortical excitatory synapsessingle neuron computationmachine learningsynaptic weights |
| spellingShingle | Toviah Moldwin Idan Segev Idan Segev Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell Frontiers in Computational Neuroscience compartmental modeling non-linear dendrites cortical excitatory synapses single neuron computation machine learning synaptic weights |
| title | Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell |
| title_full | Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell |
| title_fullStr | Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell |
| title_full_unstemmed | Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell |
| title_short | Perceptron Learning and Classification in a Modeled Cortical Pyramidal Cell |
| title_sort | perceptron learning and classification in a modeled cortical pyramidal cell |
| topic | compartmental modeling non-linear dendrites cortical excitatory synapses single neuron computation machine learning synaptic weights |
| url | https://www.frontiersin.org/article/10.3389/fncom.2020.00033/full |
| work_keys_str_mv | AT toviahmoldwin perceptronlearningandclassificationinamodeledcorticalpyramidalcell AT idansegev perceptronlearningandclassificationinamodeledcorticalpyramidalcell AT idansegev perceptronlearningandclassificationinamodeledcorticalpyramidalcell |