Neurocognitive Modeling Lessons

A central motivation for using ngc-learn is to flexibly build computational models of neuronal information processing, dynamics, and credit assignment (as well as design custom instantiations of one’s own mathematical formulations and ideas). In this set of tutorials, we will go through the central basics of using ngc-learn’s in-built biophysical components, also called “cells” and “synapses”, to craft and simulate adaptive neural systems and biophysical computational models.

Usefully, ngc-learn starts with a collection of cells – those that are partitioned into those that are graded / real-valued (ngclearn.components.neurons.graded) and those that spike (ngclearn.components.neurons.spiking). In addition, ngc-learn supports another collection called synapses – generally, those that adapt (or “learn”) with biological credit assignment building blocks (such as those in ngclearn.components.synapses.hebbian) such as spike-timing-dependent plasticity and multi-factor rules. With the in-built, standard cells and synapses in these two core collections, you can readily construct a wide variety of models, recovering many classical ones previously proposed in computational neuroscience and brain-inspired computing research (many of these kinds of models are available for external download in the Model Museum).

Sensory Input Encoding / Transformation

• Lecture 1A: Distributional Spike Train (Encoding) Models
• Lecture 1B: Trace Variables and Filtering

Spiking Neuronal Cells

• Lecture 2A: The Simplified Leaky Integrator Cell
• Lecture 2B: The Leaky Integrate-and-Fire Cell
• Lecture 2C: The FitzHugh–Nagumo Neuronal Cell
• Lecture 2D: The Izhikevich Neuronal Cell
• Lecture 2E: The Adaptive Exponential Integrator Cell
• Lecture 2E: The Hodgkin-Huxley Cell

Graded Neuronal Cells

• Lecture 3A: The Rate Cell Model
• Lecture 3B: Error Cell Models

Synapses and Forms of Plasticity

• Lecture 4A: Dynamic Synapses and Conductance
• Lecture 4B: Hebbian Synaptic Plasticity
• Lecture 4C: Spike-Timing-Dependent Plasticity
• Lecture 4D: Reward-Modulated Spike-Timing-Dependent Plasticity
• Lecture 4E: Short-Term Plasticity

Model and Analysis Tools

• Plotting and Visualization
• Metrics and Measurement Functions
• Numerical Integration
• Density Modeling and Analysis