Machine Learning

How Machine Learning Algorithms Works : An Overview

Algorithms put logic, science and reasoning behind computer machines. Finding best algorithm to use in a specific case is called experience, knowledge and need. You can call it as an another algorithm.

 

Algorithms in Machine Learning

Algorithms in Machine Learning (ML) borrows principles from computer science. There are many algorithms as no one algorithm works best for every problem,

  • How does youtube suggest you videos ?
  • Which face has what name;  How facebook knows ?
  • How google map is able find best & fastest route between office and home?

All above question has one common answer “Algorithms”. Diffrent problem will have diffrent algorithm (solution). Google map uses its own routing algorithm which is again set of many small sub algorithms.

Machine learning is a subset of Artificial Intelligence and field of study which harnesses principles of computer science & statistics. AILabPage members intuitively calls statistics as a graphical branch of mathematics. AILabPage defines machine learning  in one simple line as below.

A focal point where business, data, experience meets emerging technology and decides to work together.

 

Machine Learning – Background

Until only recently even though Machine Learning term got coined in 1950’s; It remained largely confined to academia. As on date disappointingly machine learning is becoming rocket science kind of complex subject and more accessible to developers as their tool.

Refer to post “The Exciting Evolution of Machine Learning for more details on time lines & history on machine learning.

Machine learning has evolved from Artificial Intelligence’s subset to Its own domain. It has reached to an inflection point – at least in terms of messaging. I remember in my school days as part of statistics class we were told something about AI and ML and we laughed then in the 1990s.

The need of today is a simple & easily accessible machine learning cloud service i.e. “Machine Learning as a Service” – MLaaS for every one at low cost.

 

Machine Learning Process Flow

ML instructs an algorithm to learn for itself by analysing data. Algorithms here learn a mapping of input to output, detection of patterns or by reward. The more data it processes, the smarter the algorithm gets.

In other words Machine learning algorithms “learns” from the observations. When exposed to more observations, the algorithm improves its predictive performance.

How Machine Learning Algorithms Works : An Overview

AILabPage Machine Learning Series – Machine Learning Process

 

In above diagram of-course there are a lot of iterations between step 5 to 4 & 3 and 5 to 3 & 2, which are not shown in the above diagram to keep it simple. Machine Learning is classified into three categories at high level depending on the nature of the learning and learning system.

  1. Supervised learningMachine gets labelled inputs and their desired outputs. The goal is to learn a general rule to map inputs to the output.
  2. Unsupervised learning: Machine gets inputs without desired outputs, the goal is to find structure in inputs.
  3. Reinforcement learning: In this algorithm interacts with a dynamic environment, and it must perform a certain goal without guide or teacher.

 

How Machine Learning Algorithms Work?

In one line answer is “In machine learning algorithm works by learning strategy to map input to output”.

Machine learning is pure mathematics or number game only. Algorithms works by understanding this game that underlines all algorithms. There are no flavours or taste in algorithms but simply different methods to do the same job. All the time it starts with learning a target function.

How Machine Learning Algorithms Works : An Overview.

AILabPage Machine Learning Series – Algorithms Behind Machine Learning

Target function (f) that best maps input variables to an output variable. For example for our input variables “a” we need to predict the value of “y”. In this case our function becomes y=f(a).

Can you predict value of the house (“Y”) in a location “LOCATION” with input variables like dimensions of the land in square feet, construction cost per square feet, annual rate of growth in price change etc.

After function has learnt so in future it becomes easy to predict value of  “y” for every new value of “a”. It does not get over here; there is always a need to keep error margin or so-called algorithm performance indicator (b). As its independent of input variable “a”. So our function looks like y = f(a) + b.

 

Why Learning a Function is Important

Learning a function has one sole reason and that to make correct possible predictions. Predictive modeling is mapping y = f(a) i.e predicting Y for every new value of “a”.

Functions are unknown for the model and as mentioned above their taste, flavour, shape or colors are really of no use / importance as long as it give possible accurate predictions. Because functions are unknown before algorithms learn hence the importance. Otherwise there would be no use to waste time and learn anything.

How Machine Learning Algorithms Works : An Overview

AILabPage Machine Learning Series – Machine Learning Flow

In Supervised learning algorithm jobs are to take some data with a known relationship. From above example for predicting house value where price of land etc are known and to create a model of those relationships is easy.

This is regression problem as output will be numerical but if we were to find expensive or not expensive then this would have fallen in category of classification problem.

 

Learning a Function – Approach

When to use which algorithm is a complex one to answer.  It entirely depends upon on the problem and data in hand to solve. Better to apply least 3-5 algorithms to find the best results or answer.

It is so important to try 3-5 some time even more different algorithms (This is not a law or any rule) on a given problem. Otherwise to know the best outcome before hand is impossible.

Remember its all about estimations, possibilities, prediction and forecasting so it is always best to make use of machine learning algorithm suites for approximations.

Different depictions of methods used make different assumptions about the function  that is about to learn or being learned. So whether it is linear or nonlinear we need to be careful.

 

Algorithm Performance Measure

Different machine learning types has different reasons and method so the algorithms. How well your algorithm has performed it can be measured with error margins. For example.

How Machine Learning Algorithms Works : An Overview

AILabPage Machine Learning Series – Machine Learning Algorithms Performance Dashboard

In case of supervised learning problems you can measure

  • Classification Error for classification problems – how many errors being made
  • Prediction error for regression problems – How far from real value

In case of unsupervised learning problems you can measure

  • Clustering – How loose or tight
  • Association – Confidence level of associations

 

Books & Other Material Referred

Open Internet & AILabPage (group of self-taught engineers) members hands on lab work.

 

Points to Note:

When to use which algorithm is a complex one to answer.  It entirely depends upon on the problem in hand to solve. Better to apply least 3 to find the best results and to have best answer.

All credits if any remains on the original contributor only. In the next upcoming post will talk about Recurrent Neural Networks in detail.

 

Feedback & Further Question

Do you have any questions about Deep Learning or Machine Learning? Leave a comment or ask your question via email . Will try my best to answer it.

 

SECaaS - Security as a Service Is the Next Big ThingConclusion –  We have seen in post above that in machine learning algorithms is all about learning the target function. It work to estimate the mapping function (f) of output variables (y) given input variables (a), or Y=f(a). Google, Match.com, and Facebook matching algorithms are examined to see more clearly how they work enthymematically.

We have also learned in the post that different machine learning algorithms make different assumptions. When to use which algorithm is a complex one to answer. So it is always best to make use of machine learning algorithm suites for approximations.

 

============================ About the Author =======================

Read about Author at : About Me

Thank you all, for spending your time reading this post. Please share your opinion / comments / critics / agreements or disagreement. Remark for more details about posts, subjects and relevance please read the disclaimer.

FacebookPage    ContactMe      Twitter

====================================================================

 

Advertisements

2 replies »

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.