From Human Neurons to Artificial Neurons continuation





After Simple Neuron and Firing rule remaining rules are below :

3. Pattern Recognition:

After simple neuron, firing rules and important application of neural networks is pattern recognition. Pattern recognition can be enforced by using a pro-act neural network that has been trained accordingly during training the network is trained to associate outputs with input patterns. When the network is used it identifies the input pattern and tries to output the associated output pattern. The power of neural networks comes to life when a pattern that has no output associated with it, is given as an input. In this case, the network gives the output that corresponds to a trained input pattern that is least different from the given pattern.

Above example is trained to recognize the patterns T and H. The associated patterns are all black and all white respectively.

Input                          Output                   Input                  Output

Above white squares represent with 1  and black squares represent with 0 then the truth tables for the 3 neurons after generalizations are below truth table.

Top Neuron:

Middle Neuron :

Bottom Neuron:

From the above tables, it can be seen the following associations can be extracted




Input                               Output

In this case, it is obvious that the output should be all blacks since the input pattern is almost the same as the “T” pattern.

Input                     Output

In this case, it is obvious that the output should be all whites since the input pattern is almost the same as the ‘H’ Pattern.

Input                                        Output

Above case, the top row is 2 errors away from the T and 3 from an H. So the top output is black. The middle row is 1 error away from both T and H so the output is random.

The bottom row is 1 error away from T and 2 away from H. Therefore the output is black. The total output of the network is still in favor of the T shape.

4. A more complicated Neuron:

The most sophisticated neuron is the McCulloch and Pitts model. It is a variety from the remaining model is that the inputs are ‘weighted’ the effect that each input has at decision making is dependent on the weight of the particular input.

From Human Neurons to Artificial Neurons





To easy to understand Human neurons to Artificial Neurons is a little bit tough but we conduct these neural networks by first trying to conclude the essential features of neurons and their internal connections in Artificial Intelligence. Then typically program to the computer to replicate these characteristics. However, because our knowledge of neurons is insufficient and our computing power is limited, our models are necessarily gross idealizations of real networks of neurons.

The Neuron model:

An Engineering Approach

1.A Simple Neuron:

In a simple neuron, Artificial Neuron is a device with many inputs but only one output. The neuron has different modes of operations in a simple neuron. One is training mode and another one is user mode. Basically the training mode, the neuron can be trained to fire, for a particular input pattern. And the user mode when a taught input pattern is detected at the input, its related to output becomes the current output in Artificial neuron. If the input pattern does not belong in the taught list of input patterns the firing rule is used to determine whether to fire or not in a simple neuron.

2.Firing rules:

In an Artificial Intelligence, the firing rules is a most important concept in neural networks and account for their high adaptability. A firing rule verifies a neuron should fire for any input pattern. Firing rules understand to all the input patterns not only the ones on which the node was trained in Artificial neurons.

A simple firing rule can be performed by using the Hamming distance technique.

In simple firing rule can take a collection of training patterns for a node, some of which generate it to fire and others which intercept it from doing so then the patterns not in the collection cause the node to fire if, on the comparison, they have more input elements in common with the nearest pattern in the 1 – taught set than with the nearest pattern in the 0 – taught set. If there is a tie then the pattern remains in the undefined state.

Example: In firing, rule take 3 – input neuron is trained to output 1 when the input (X1, X2, and X3)  is 101 or 111 and to output 1 when the input is 000 or 001 and to output is 0 the final output  truth table below is:

In the above example of the way the after applying Firing, a rule is to take the pattern 010. Firing rule differs from 000 in 1 element, from 001 in 2 elements, from 101 in 3 elements and from 111 in 2 elements. Therefore, the close the pattern is 000 which belongs in the 0 – taught set. It necessary that the neurons do not fire when the input is 001, on the other hand, is equal distance from two trained patterns that have different outputs and consequently the output stays undefined 0/1.




For more the difference between the two truth tables is called the generalization of the neuron. The firing rule gives the neuron a sense of similarity and authorizes it to respond sensibly to patterns not seen during training.

Neural networks overview

NeuralNetworks:

An Artificial Neural Network(ANN) is an information processing model that is quality, by the way, the biological nervous system, such as the brain, process information. The key element of this model is the establishing structure of the information processing system.




A Neural network is a group of a huge number of highly interconnected processing elements or neurons.

In Artificial Neural Network is Connectionist systems are computing systems are inspired by the biological neural networks that constitute brains.

For example in image recognition, they might learn to identify images that contain cows by analyzing example images that have been manually labeled “cow” or “no cow” and using the results to identify cows in other images.

Why use neural networks?

Neural networks, with their extraordinary ability to obtain meaning from complicated or lose data, can be used to extract patterns and detect trends that are too complex to be noticed by either computer or humans.

A trained neural network can be thought of as an “authority” in the category of information it has been given to analysis. This authority can be used to provide estimate given new situations of interest and answer “what if ” questions.

Advantages:

Adaptive learning: The capacity to learn how to do tasks based on the data given for instructions or initial experience.

Self Organization: An Artificial Neural Network can create its own organization or representation of the information it receives during learning time.

Real-Time Operation: An Artificial Neural Network can be carried out in parallel, and special hardware devices are planning and make which take a lead of this ability.




Summary: An Artificial Neural Network is an information processing, recognition to identify images. And processing elements like neurons for complicated problems. In ANN have more advantages including adaptive learning, self-organization, fault tolerance and real-time operations for the neural network.

Skewness and Kurotsis

You have an idea of data then will go with the Shape of data.




The shape of data measured by

1.Skewness

2.Kurtosis

1.Skewness (Symmetry):

Measures look at how lopsided distributions are how far from the ideal of the normal curve they are when the median and the mean are different, the distribution is skewed. The huge difference is the difference skew.

If the skewness is extreme, the researcher should either transform the data to make them better resemble a normal curve or else use a different set of statistics

Indicators of Skewness:

Frequency curve is not symmetrical bell-shaped and the sum of positive deviation is not equal to the sum of negative deviation

Finally, it measures of asymmetry of data.

Positive or right skewed: Longer right tail

Negative or left skewed: Longer left tail

Let x1, x2, x3…..xn be n observations. Then

Kurtosis :

It is a measure of the “taildeness” of the probability distribution of a real-valued random variable. And it is concerned with the degree of Flatness in a curve.

Kurtosis is a shape of a probability distribution and, just as for skewness, there are different ways of quantifying it for a theoretical distribution and corresponding ways of estimating it from a sample from a population.

These three are Kurtosis prototype shapes.



1. Leptokurtic: A curve which is more peaked than the normal

2. Mesokurtic: A normal curve is called as a mesokurtic curve.

3. PlatyKurtic: A flat curve than normal is called platykurtic.

Kurtosis Formula:

Let x1,x2,x3…xn be n observations. Then,

Outlier in Statistics




Outlier:

An outlier is any value that is numerically faraway from most of the other data points in a set of data. And an outlier is not uncommon to find an outlier in a data set.

I) In statistics, an outlier is an observation point that is faraway from remaining observations.

II) An outlier may be due to a lack of consistency in the measurement

III) It may be cause serious problems in statistical analysis and can occur in any distribution.

Outliers can have many causes such as:

A)Measurement or input error

B)Data corruption

C)Outlier values, possibly in the context of non – outlier values to see if there are a systematic relationship.

Two activities are essential for characterizing a set of data:

1. The overall shape of the graphed data for important features, including coordination and departure from acceptance.

2.The data for the different examination that is far removed from the mass of data. Here two graphical techniques for identifying outliers, scatter plots and box plots, along with an analytic procedure for detecting

Measures of Central Tendency(Mode, Median, Mean)




Averages:

Mode:

Mode is a most often occurring value in a distribution .

I)The most numerous category

II)For ratio data, often implies that data have been grouped in some way. It can be more or less created by the grouping procedure

III)For theoretical distributions – simply the location of the peak on the frequency distribution.

Advantages :

I)Quick and easy to compute

II)Unaffected by extreme scores unrepresentative

III)Can be used at any level of measurement

Disadvantages:

I)Terminal statistics

II)A given sub group cloud make this measure

Mean:

Arithmetic average the sum of all values in a distribution divided by the number of cases

I) Center of gravity

II)Even partitions the sum of all measurement among all cases, an average of all measures

Advantages and Disadvantages:

I) Crucial for inferential statistics

II) Mean is not very resistant to outliers

III) A “trimmed mean” may be better for descriptive purposes

R: mean(x)

Arithmetic mean, Harmonic mean, Geometric mean, also f- mean, truncated mean, power mean, weighted mean, etc.

Median :

midpoint in the distribution below which half of the cases reside.

I)less useful for inferential purposes

II)more resistant to the effects of outliers.

The mean or average of data values is

mean = sum of all data values/ number of data values

Statistics in Artificial Intelligence




Statistics :

The science of collecting, describing and interpreting data.

Important note on Statistics:

Many studies generate large number of data points and to make sense of all that data researchers use statistics that summarize the data providing a better understanding of overall tendencies within the distributions of scores.

It is a branch of mathematics dealing with the collection, analysis, explanation presentation and organization of data.

Statistics deals with all aspects of data including the planning of data collections in terms of the design of surveys experiments.

Two main statistical methods are used in data analysis descriptive statistics which summarize data from a sample using indexes such as the mean or standard deviation

We use statistics for many reasons:

To mathematically describe

To draw conclusions from our results

To test hypotheses

To test for relationships among variables

Helps us recognize underlying trends and tendencies in the data.

Two types of statistics :

Numerical representation of data basically divided into two types

Descriptive Statistics: Collection, Presentation and summarize data




Inferential Statistics : Making decisions and drawing conclusions about populations

Descriptive Statistics :

It is a summary statistics that quantitatively describes a collection of information and analyzing the data. Summarize a sample data, rather than use data learn about the population that the sample of data is thought to represent.

I)Measures of central tendency:

It means that the central position of the frequency distribution for the group for data.

II)Measures of dispersion:

It means that the process of distributing data at a certain speed.

III)Measures of symmetry:

It means that symmetric distribution of a group of data. Like positively skewed and negative skewed distribution.

2.Inferential:

Test for the significant difference between groups / significant relationships among variables within the sample.

I) t – ratio, chi-square, and beta – value within a group

II)It is information about your immediate group of data.

Summary: Statistics are mainly mathematically described, descriptive types and inferential types of statistics in the real-time world.

Why is Artificial Intelligence more important?




AI adds intelligence :

Artificial Intelligence to existing products. In most of the cases, AI capabilities, for example, SIRI was added as a feature to a new production of Apple Products. And AI Chatbots are servicing purpose example below like this for help in some applications.

AI automates repetitive learning :

AI is different from hardware – driven, robotic automation, instead of manual testing and it performs high – volume and very frequent for tasks.

Artificial Intelligence  Analyzes  and Deeper data :

In AI analyzes need more data using a neural network that has hidden layers. In the case of building a fraud detection system with hidden layers are impossible for a few years ago. Artificial Intelligence has changed with incredible in Big data technology here we need lot data to train deep learning because they learn directly from the Big Data.

AI achieves Accuracy :

Artificial Intelligence has incredible accuracy though in Deep Learning. For example, Google search using Alexa for interaction based on deep learning. And Visualization using Google Car etc.

How Artificial Intelligence is being used:

Nowadays every industry has a high demand for AI capabilities like medical research, legal assistance, and risk notification

In Health Care Domain:

Artificial Intelligence application can be provided to customize medicine and X-ray readings. Personal health care assistants can act as life coaches, exercise.

Manufacturing:

AI can analyze Internet of Things data as it streams from connected equipment to forecast load and demand using deep learning and network used with related data.

Retail:

Artificial Intelligence provides virtual shopping capabilities that offer personalized discuss purchase offers with the consumer.




Sports:

Artificial Intelligence is used to capture images of gameplay provide coaches with reports on how to good organize the game. Set and optimizing field positions and strategy.

Summary: Nowadays AI more useful and needy technology in every industry.

Artificial Intelligence (AI)

What is Artificial Intelligence?

Artificial Intelligence is the science of making machines to do things that would require intelligence of done by men – Marvin Minsky

Artificial Intelligence Aspects:

Artificial Intelligence is a Planning, Vision(image recognize), Speech to text – Text to Speech, Robotics, Machine Learning(Deep Learning and Natural Language)




AI Planning :

Planning is a key ability for an intelligent system for flexibility through the construction of a sequence of actions to achieve their goals. And it involves the representation of actions and different types of models and techniques for efficiently searching the space of possible plans.

AI Robotics:

In the early days of AI and Robotics were strongly connected. AI consisted of building complete intelligent robots. The special issue will collect in a single place a set of reports that are innovative and mature inter disciplinary work in the integration of AI and Robotics.

AI Machine Learning :

Machine Learning is a particular approach to artificial intelligence and ML use very little rule-based systems.  Nowadays most of the applications are indeed will use ML, on the other hand, Deep Learning.

AI Natural Language Processing (NLP):

Artificial Intelligence systems can understand, interpret and communicate in human language. For example in NLP like Chat Bot for conversational purpose.

AI Vision :

A machine can see objects around like a human. The technology is used for an application like robot guidance

ex: Google Car

AI Speech Analytics :

It is the ability to process speech or convert written information to speech it means that interactions with machines

Example: SIRI, Google Maps.




What is Learning?

Learning is the Pattern/ relations in the data and learning requires training data on which a system trained.

The system needs training and the system learns from data.

Artificial Intelligence is the new UI :

1. Chatbots help people access the knowledge repositories in natural language.

2. The user experience like talking to another human being.