The gradient clusteron: A model neuron that learns to solve classification tasks via dendritic nonlinearities,
Toviah Moldwin1, Menachem Kalmenson2, Idan Segev1,2
1Edmond and Lily Safra Center for Brain Sciences, the Hebrew University of Jerusalem, Jerusalem, Israel.
Plos Computational Biology
|May 24, 2021
Summary
The gradient clusteron (G-clusteron) algorithm enhances neural network pattern recognition by adjusting synapse locations and weights. This structural and functional plasticity improves classification accuracy on tasks like MNIST.
Area of Science:
- Computational neuroscience
- Machine learning
- Artificial intelligence
Background:
- Synaptic clustering on neuronal dendrites is crucial for pattern recognition.
- Existing models like the clusteron use structural plasticity inspired by NMDA receptors.
Purpose of the Study:
- To introduce the gradient clusteron (G-clusteron) model for enhanced classification.
- To investigate both structural and functional plasticity in synaptic organization.
Main Methods:
- Developed an analytically-derived gradient descent rule for synapse attraction/repulsion.
- Tested the G-clusteron on the MNIST handwritten digit dataset.
- Derived and applied synaptic weight update rules (functional plasticity).
Main Results:
- G-clusteron achieved ~85% accuracy on the MNIST task (all-versus-all) with softmax activation.
- Functional plasticity alone improved accuracy to ~89% on MNIST.
- Combined plasticity rules enabled solving the XOR problem from arbitrary initial conditions.
Conclusions:
- The G-clusteron model offers an effective approach to synaptic organization for classification tasks.
- Both structural and functional plasticity contribute to improved learning performance.
- This biologically-inspired model shows promise for advancing artificial neural networks.
Related Concept Videos
01:21Electrochemical Gradient and Channel Proteins: An Overview
3.6K
An electrochemical gradient is a fundamental concept in biology and chemistry. It regulates the movement of ions across cell membranes. This movement is influenced by two factors:
The electrical gradient: The electrical gradient across cell membranes refers to the difference in electric charge between the inside and outside of a cell. This difference drives the movement of ions towards or away from the cells. For instance, if the inside of the cell is more negatively charged relative to...
The electrical gradient: The electrical gradient across cell membranes refers to the difference in electric charge between the inside and outside of a cell. This difference drives the movement of ions towards or away from the cells. For instance, if the inside of the cell is more negatively charged relative to...
3.6K
01:01Neuroplasticity
1.0K
Neuroplasticity reflects the brain's remarkable capacity to adapt and evolve, responding dynamically to learning, experiences, or injury by reorganizing its neural circuitry. This reorganization involves creating new neural connections and refining old ones through a series of biological processes that contribute to the brain's lifelong development and adaptability.
1.0K
01:26What is an Electrochemical Gradient?
122.4K
Adenosine triphosphate, or ATP, is considered the primary energy source in cells. However, energy can also be stored in the electrochemical gradient of an ion across the plasma membrane, which is determined by two factors: its chemical and electrical gradients.
The chemical gradient relies on differences in the abundance of a substance on the outside versus the inside of a cell and flows from areas of high to low ion concentration. In contrast, the electrical gradient revolves around an...
The chemical gradient relies on differences in the abundance of a substance on the outside versus the inside of a cell and flows from areas of high to low ion concentration. In contrast, the electrical gradient revolves around an...
122.4K
01:19Graded Potential
5.8K
Graded potentials are localized fluctuations in the cell membrane's electrical charge, commonly found in the dendrites of neurons. The magnitude of these potential changes depends on the strength of the initiating stimulus. In a membrane at its resting potential, a graded potential signifies a voltage shift either above -70 mV or below -70 mV.
Graded potentials fall into two categories: depolarizing and hyperpolarizing. Depolarizing graded potentials typically occur when sodium (Na+) or...
Graded potentials fall into two categories: depolarizing and hyperpolarizing. Depolarizing graded potentials typically occur when sodium (Na+) or...
5.8K
01:19The Role of Ion Channels in Neuronal Computation
3.4K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.4K
01:25Neural Circuits
2.0K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
2.0K