[NEST Simulator] Classification and summarization of NEST example

NEST Simulator officially provides 35 example network, the number is not many, so precious

In this paper, these example are classified for follow-up study.

Classification according to 2 main lines:
1.example content: Includes what example itself expects to provide, and what concerns I have about it.
2.example Priority: This is the sequence to follow when you start learning about example.

Example Content

Device usage (DU): The use of nest device is simulator basis.
Neuron behavior (NB): Neuron theory and behavior, belong to computational neuroscience content
Synapse Behavior (SB): Synapse theory and behavior, belong to computational neuroscience content
Modeling Methods (MM): Focus on understanding the modeling method, belongs to the simulator application.
Extension (EX): extended knowledge to broaden your horizons, with little chance of engineering

Example Priority level　　

Class 0: Basic methods and knowledge that directly impact modeling
Class 1: Specific knowledge that affects model complexity
Class 2: Advanced methods for increased efficiency
Class 3: Extensibility methods and knowledge that do not directly model impact
Class 0,1 affects the model, Class2, does not affect the model, but affects efficiency

The following are all example List table:

Index	Example	Description	Class	Priority of Class
1	Balanced Neuron Example	Content: excitatory & inhibitory neurons populations issued Poisson spike train to the same 1 neuron, Find one of the neuron firing rate so that it is the same as excitatory firing rate Attention point: Matching method of firing rate between different neurons population	Mm	0
2	Campbell & Siegert Approximation Example	Content: Multiple Poisson generators use different rate and weight driver if neuron make it firing, and compare its firing rate with Iaf_psc_alpha model. Focus: Modeling methods, non-comparison results	Mm	0
3	Compare Tsodyks-markram Synapse Models	-	-	-
4	Comparing precise and grid-based neuron models	Content: Compare the differences between precise and grid-based neuron models Focus: Not concerned about precise, so the example is not very meaningful	EX	3
5	Correlospinmatrix_detector Example	Content: Binary neuron model and difference comparison. Ginzburg_neuron and Mcculloch_pitts_neuron are binary neuron, Correlospinmatrix_detector Comparing binary differences Focus: Binary model has great potential for no attention, But if it is found to be effective later, That will have a great impact on neuromorphic system design. This is a potential point of eruption for blockbuster	EX	3
6	Example for the Quantal_stp_synapse	Description: Describes the synapse model of STP (including short term depression/shor term facilitation). Describe the change of weight from the release site level, Mechanism may be the basis for neuromorphic hardware, However, it is possible that the hardware is not directly implemented. Focus: Model itself and STP rules	SB	1
7	Example of the Adaptive exponential integrate and Fire (AdEx) in NEST.	Content: Two different implementations of aeif_cond_alpha comparing speed and accuracy Concern: the example is not concerned with the comparison process and results. Aeif_cond_alpha itself needs to be understood later, This model uses alpha-function to model synaptic conductance, If Alpha-function really can have an impact on ng/fng, Then consider adding it to the hardware model. Introduction to the Aeif_cond_alpha see https://jn.physiology.org/content/94/5/3637	NB	1
8	Example of the Tsodyks2_synapse in NEST	Content: Tsodyks2_synapse's Simple network The point of concern: the network itself is nothing special, Tsodyks2_synapse model needs to learn, Related to STP	SB	1
9	IAF Neuron Example	Content: IAF Neuron test Focus: basic example, need to master	NB	0
10	IF Curve Example	Content: Enter the incremental current into the neurons in turn, Observe the relationship of Current-spiking frequence. Focus: Basic, need to master	NB	0
11	Initial membrane voltage	Content: In the absence of external input, IAF_COND_EXP_SFA_RR model Membrain potential tend to -70mv The point of concern: the cause of the phenomenon is essentially an understanding of neural dynamics. (may be reseting potential, but because there are multiple membrane potential for -70mv, it is not OK)	NB	1
12	Intrinsic currents spiking	Content: Refinement to the receptor category level, temporarily do not see, temporarily do not consider this level Focus: None	EX	3
13	Intrinsic Currents subthreshold	Content: Intrinsic current related to the temporary don't see Focus: None	EX	3
14	Multi-compartment Neuron Example	Content: multi-compartment neuron test. Focus: Multi-compartment model divides neuron dendrite into three types: distal, proximal, Soma, Each synapse can receive current and spike, and different positions of dendrite have different effects on membrane potential, This model is temporarily ignored, If the late appearance of model is likely to be useful, then focus on the Multi-comparment model.	EX	3
15	Multimeter to file example	Content: Use of multimeter Focus: Basic, to Master	DU	0
16	One neuron example	Content: Basics Focus: Basic, to Master	Mm	0
17	One_neuron_with_noise	Content: Basics Focus: Basic, to Master	Mm	0
18	Plot Weight matrices Example	Content: Extracts all the connection strength of all types between the 2 neuron group to be saved in the matrix. and analysis and visualization Focus: Basic, to Master	Mm	0
19	Pulse Packet Example	Content: Membrain Potential status changes after receiving pulse package (Gaussian distributed transient spike volley) Focus: Focus on the generation of spiking	Mm	0
20	Random Balanced Network (alpha synapses) connected with NEST	Content: Implement random connection Create network with Nest's connection roution Focus: Basic, to Master	Mm	0
21st	Random Balanced Network (alpha synapses) connected with NumPy	Content: Using NumPy to implement the random connection create network Focus: Basic, to Master	Mm	0
22	Random Balanced Network (delta synapses)	Content: Create a network based on a paper of Brunel N Focus: How to build a basic network	Mm	0
23	Random Balanced Network (exp synapses, multiple time constants)	Content: Based on the paper of Brunel N, Synapse produces exponential PSP, The receptor type defines the time Constant,receptor types that corresponds to the PSP evenly distributed. Focus: All with receptor type is not considered, Because the first thing to do now is to understand Neuron group from the network level, Walk through the flow and then carefully understand the neuron model. But I also think that receptor type can be introduced into the hardware model to add variability.	NB	1
24	Repeated stimulation	Content: generates stimulus within a specified time interval, using ' origin ' to produce stimulus repeatedly Focus: Basic, to Master	Mm	0
25	Sinusoidal Poisson Generator Example	Content: Introduction to the use of Sinusoidal_poisson_generator Focus: The use of basic device, to master	DU	0
26	Sinusoidal_gamma_generator	Content: Introduction to the use of Sinusoidal_gamma_generator Focus: The use of basic device, to master	DU	0
27	Spike synchronization through Subthreshold oscillation	Content: reproduce "simple Networks for spike-timing-based computation, with application to olfactory Processing.pdf "in the phenomenon, Spike synchronization behavior under the condition of subthreshold oscillation Focus: FNG, focus on generating FNG modeling methodology, in order to develop new FNG	NB	1
28	Structural plasticity Example	Description: This is an interesting example. 2 groups of population, no connection at first, Using the structural plasticity model in buts, M., & Von Ooyen A. (2013), You can create and delete synapses dynamically in the simulation process until you reach the desired electrical activity Focus: This synapse creates a method that differs from the simple adjustment of weight I understand before, but rather the process of synapse creation and disappearance, Careful understanding and use of this process may significantly reduce the memory capacity of the synpase.	NB	1
29	Test of the adapting exponential integrate and fire model in nest	Content: test for adaptive integrate and Fire model (ADEX) Focus: The idea of learning the test process, In addition the model requires learning (Brette and Gerstner (2005) J. Neurophysiology and reproduces Figure 2.C)	NB	1
30	Test of the adapting exponential integrate and fire model in nest	Content: Ibid., for (Brette and Gerstner (2005) J. Neurophysiology and reproduces Figure 3.D) Focus: Ibid.	NB	1
31	Tsodyks depressing example	Content: Depressing Tsodyks Synapse example, verify its behavior, see "Neural Networks with Dynamic synapses.pdf" Focus: The paper is important and should be looked at carefully.	SB	1
32	Tsodyks Facilitating Example	Content: Ibid., change parameter formation facilitating Tsodyks synapse Focus: Ibid.	SB	1
33	Twoneurons	Content: 2 Neuron connections Focus: Basic, need to master	Mm	0
34	Using CSA for connection setup	Content: Describes an efficient way to create network connection independent of simulator: Using Connection Set Algebra (CSA), See "Efficient generation of connectivity in neuronal networks.pdf" Focus: Mastering This approach is an advanced way of using tools	Mm	2
35	Using CSA with topology layers	Content: IBID., create topology layer between layers with CSA connection Focus: Ibid.	Mm	2
Note: [1]. The above concerns are based on the neuromorphic angle, not the computational neuroscience point of view. The direct representation is to ignore the model details of some Synapse/neuron models. [2]. Example 3 omitted, because example did not describe, and according to code obviously not unimportant [3] ng/fng:neurogroup/functional Neurogroup

Below is the example priority list, which is listed according to table above, containing the ordering within each class, that is, the learning order:

Class 0:
#neuron Connection
IAF Neuron Example
One neuron example
One_neuron_with_noise
Twoneurons
IF Curve Example

#network
Balanced Neuron Example
Random Balanced Network (alpha synapses) connected with NEST
Random Balanced Network (alpha synapses) connected with NumPy
Random Balanced Network (delta synapses)

#device
Sinusoidal Poisson Generator Example
Sinusoidal_gamma_generator
Pulse Packet Example
Multimeter to file example

#modeling methods
Repeated stimulation
Plot Weight matrices Example
Campbell & Siegert Approximation Example

Class 1:
#synapse
Example for the Quantal_stp_synapse
Example of the Tsodyks2_synapse in NEST
Tsodyks depressing example
Tsodyks Facilitating Example
Structural plasticity Example

#NG
Spike synchronization through Subthreshold Oscillation:ng
Test of the adapting exponential integrate and fire model in Nest:new model
Test of the adapting exponential integrate and fire model in Nest:new model
Example of the Adaptive exponential integrate and Fire (AdEx) in NEST.: Algorithm Implementation
Initial Membrane Voltage:no Use
Random Balanced Network (exp synapses, multiple time constants): receptor type

Class 2:
Using CSA for connection setup
Using CSA with topology layers

Class 3:
Comparing precise and grid-based neuron models
Correlospinmatrix_detector Example
Intrinsic currents spiking
Intrinsic Currents subthreshold
Multi-compartment Neuron Example

[NEST Simulator] Classification and summarization of NEST example