microsoft/AI-For-Beginners

Public

mirrored fromhttps://github.com/microsoft/AI-For-BeginnersAvailable

CodeCommitsIssuesPull requestsActionsInsightsSecurity
1bcb2489ab4af0973915b702e09175795612e4b0

Branches

Tags

  • No tags available.
0Branches0Tags
Go to file
Add file
Code

Clone

HTTPS

Download ZIP

lessons/3-NeuralNetworks/README.md

48lines · modecode

1# Introduction to Neural Networks
2
3![Summary of Intro Neural Networks content in a doodle](../sketchnotes/ai-neuralnetworks.png)
4
5As we discussed in the introduction, one of the ways to achieve intelligence is to train a **computer model** or an **artificial brain**. Since the middle of 20th century, researchers tried different mathematical models, until in recent years this direction proved to be hugely successful. Such mathematical models of the brain are called **neural networks**.
6
7> Sometimes neural networks are called *Artificial Neural Networks*, ANNs, in order to indicate that we are talking about models, not real networks of neurons.
8
9## Machine Learning
10
11Neural Networks are a part of a larger discipline called **Machine Learning**, whose goal is to use data to train computer models that are able to solve problems. Machine Learning constitutes a large part of Artificial Intelligence, however, we do not cover classical ML in this curricula.
12
13> Visit our separate **[Machine Learning for Beginners](http://github.com/microsoft/ml-for-beginners)** curriculum to learn more about classic Machine Learning.
14
15In Machine Learning, we assume that we have some dataset of examples **X**, and corresponding output values **Y**. Examples are often N-dimensional vectors that consist of **features**, and outputs are called **labels**.
16
17We will consider the two most common machine learning problems:
18
19* **Classification**, where we need to classify an input object into two or more classes.
20* **Regression**, where we need to predict a numerical number for each of the input samples.
21
22> When representing inputs and outputs as tensors, the input dataset is a matrix of size M×N, where M is number of samples and N is the number of features. Output labels Y is the vector of size M.
23
24In this curriculum, we will only focus on neural network models.
25
26## A Model of a Neuron
27
28From biology, we know that our brain consists of neural cells (neurons), each of them having multiple "inputs" (dendrites) and a single "output" (axon). Both dendrites and axons can conduct electrical signals, and the connections between them — known as synapses — can exhibit varying degrees of conductivity, which are regulated by neurotransmitters.
29
30![Model of a Neuron](images/synapse-wikipedia.jpg) | ![Model of a Neuron](images/artneuron.png)
31----|----
32Real Neuron *([Image](https://en.wikipedia.org/wiki/Synapse#/media/File:SynapseSchematic_lines.svg) from Wikipedia)* | Artificial Neuron *(Image by Author)*
33
34Thus, the simplest mathematical model of a neuron contains several inputs X<sub>1</sub>, ..., X<sub>N</sub> and an output Y, and a series of weights W<sub>1</sub>, ..., W<sub>N</sub>. An output is calculated as:
35
36<img src="images/netout.png" alt="Y = f\left(\sum_{i=1}^N X_iW_i\right)" width="131" height="53" align="center"/>
37
38where f is some non-linear **activation function**.
39
40> Early models of neuron were described in the classical paper [A logical calculus of the ideas immanent in nervous activity](https://www.cs.cmu.edu/~./epxing/Class/10715/reading/McCulloch.and.Pitts.pdf) by Warren McCullock and Walter Pitts in 1943. Donald Hebb in his book "[The Organization of Behavior: A Neuropsychological Theory](https://books.google.com/books?id=VNetYrB8EBoC)" proposed the way those networks can be trained.
41
42## In this Section
43
44In this section we will learn about:
45* [Perceptron](03-Perceptron/README.md), one of the earliest neural network models for two-class classification
46* [Multi-layered networks](04-OwnFramework/README.md) with a paired notebook [how to build our own framework](04-OwnFramework/OwnFramework.ipynb)
47* [Neural Network Frameworks](05-Frameworks/README.md), with these notebooks: [PyTorch](05-Frameworks/IntroPyTorch.ipynb) and [Keras/Tensorflow](05-Frameworks/IntroKerasTF.ipynb)
48* [Overfitting](05-Frameworks#overfitting)
49