Self-supervised learning of probabilistic prediction through synaptic plasticity in apical dendrites: A normative model

Rao, A. Ravishankar; Legenstein, Robert; Subramoney, Anand; Maass, Wolfgang

doi:10.1101/2021.03.04.433822

Cited by 3 publications

(4 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These works however did not consider synaptic clustering in the context of associative learning as we did. The work by Rao et al ( 2022 ) developed a plasticity rule termed Dendritic Logistic Regression that enables the apical compartment to predict somatic activity. This objective is somewhat related to our association objective, although we did not attempt to produce a precise probabilistic prediction as in Rao et al ( 2022 ).…”

Section: Discussionmentioning

confidence: 99%

“…The work by Rao et al ( 2022 ) developed a plasticity rule termed Dendritic Logistic Regression that enables the apical compartment to predict somatic activity. This objective is somewhat related to our association objective, although we did not attempt to produce a precise probabilistic prediction as in Rao et al ( 2022 ). The authors of this paper also did not consider a clustering objective, hence activity in their model for a given apical activation is distributed over all branches.…”

Section: Discussionmentioning

confidence: 99%

“…These works however did not consider synaptic clustering in the context of associative learning as we did. The work by Rao et al (2022) developed a plasticity rule termed . /fnins.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Context association in pyramidal neurons through local synaptic plasticity in apical dendrites

Baronig,

Legenstein

2024

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

The unique characteristics of neocortical pyramidal neurons are thought to be crucial for many aspects of information processing and learning in the brain. Experimental data suggests that their segregation into two distinct compartments, the basal dendrites close to the soma and the apical dendrites branching out from the thick apical dendritic tuft, plays an essential role in cortical organization. A recent hypothesis states that layer 5 pyramidal cells associate top-down contextual information arriving at their apical tuft with features of the sensory input that predominantly arrives at their basal dendrites. It has however remained unclear whether such context association could be established by synaptic plasticity processes. In this work, we formalize the objective of such context association learning through a mathematical loss function and derive a plasticity rule for apical synapses that optimizes this loss. The resulting plasticity rule utilizes information that is available either locally at the synapse, through branch-local NMDA spikes, or through global Ca2+events, both of which have been observed experimentally in layer 5 pyramidal cells. We show in computer simulations that the plasticity rule enables pyramidal cells to associate top-down contextual input patterns with high somatic activity. Furthermore, it enables networks of pyramidal neuron models to perform context-dependent tasks and enables continual learning by allocating new dendritic branches to novel contexts.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Context association in pyramidal neurons through local synaptic plasticity in apical dendrites

Baronig,

Legenstein

2024

Front. Neurosci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The synaptic plasticity rule utilized in (Legenstein and Maass, 2011) and (Limbacher and Legenstein, 2020) is spike time dependent plasticity rule. In a recent study, Rao et al (2022) employ Dendritic Logistic Regression to define connection weights. Furthermore Moldwin and Segev (2020) used perceptron rule to define synaptic connection weights.…”

Section: Introductionmentioning

confidence: 99%

Improving a cortical pyramidal neuron model’s classification performance on a real-world ecg dataset by extending inputs

et al. 2023

View full text Add to dashboard Cite

Pyramidal cells are the most prevalent neuronal type in the cortex, receiving thousands of synaptic inputs from all over the brain, and sending the largest axon outputs. They have a variety of active conductivities and complex morphologies that support highly nonlinear dendritic calculations. There has been a growing interest in understanding the classification abilities of pyramidal neurons. The perceptron learning algorithm, one of the foundations of machine learning, uses the highly simplified mathematical abstraction of a neuron, and it is unclear to what extent real biophysical neurons can perform perceptron like learning. In this article, we investigated the performance of a pyramidal neuron model in the classification problem of a two-class ECG dataset for different synaptic regions by using the perceptron learning method. The main purpose of this study is to reveal what role the soma, basilar and apical dendrites play in a classification problem. We concluded that when the synaptic receptor locations are selected close to the soma, classification performance close to Perceptron Learning in a Cortical Pyramidal Neuron Model the single layer perceptron can be obtained. The results indicated that the pyramidal neuron can successfully classify real-world data.

show abstract

Classification of a Two-Class ECG Dataset Based on Perceptron Learning in a Cortical Pyramidal Neuron Model

Bozkır

Özcan

Köse

et al. 2022

Preprint

View full text Add to dashboard Cite

Pyramidal cells are the most prevalent neuronal type in the cortex, receiving thousands of synaptic inputs from all over the brain, and sending the largest axon outputs. They have a variety of active conductivities and complex morphologies that support highly nonlinear dendritic calculations. There has been a growing interest in understanding the classification abilities of pyramidal neurons. The perceptron learning algorithm, one of the foundations of machine learning, uses the highly simplified mathematical abstraction of a neuron, and it is unclear to what extent real biophysical neurons can perform perceptron like learning. In this article, we investigated the performance of a pyramidal neuron model in the classification problem of a two-class ECG dataset for different synaptic regions by using the perceptron learning method. The main purpose of this study is to reveal what role the soma, basilar and apical dendrites play in a classification problem. We concluded that when the synaptic receptor locations are selected close to the soma, classification performance close to the single layer perceptron can be obtained. The results indicated that the pyramidal neuron can successfully classify real-world data.

show abstract

Self-supervised learning of probabilistic prediction through synaptic plasticity in apical dendrites: A normative model

Cited by 3 publications

References 34 publications

Context association in pyramidal neurons through local synaptic plasticity in apical dendrites

Context association in pyramidal neurons through local synaptic plasticity in apical dendrites

Improving a cortical pyramidal neuron model’s classification performance on a real-world ecg dataset by extending inputs

Classification of a Two-Class ECG Dataset Based on Perceptron Learning in a Cortical Pyramidal Neuron Model

Contact Info

Product

Resources

About