Artificial Intelligence vs Machine Learning

 Difference between Artificial Intelligence and Machine Learning, a roadmap for achieving Artificial Intelligence.

                                                               

ML vs AI

Machine Learning is a concept under AI where we let the machine learn things on its own if it is provided with sufficient data and computational power(the machines are not explicitly programmed). Here machines learn like humans by gaining experiences using thinking capabilities, where experiences are nothing but information or data and thinking capabilities are nothing but computational power.

Artificial Intelligence is a much broader concept where machines are capable of achieving things that we human beings consider as “smart”. There are multiple ways to achieve AI, but the best and the most groundbreaking discoveries are happening in the field of machine learning

It could be said that now the major driving force of AI is Machine Learning which is the reason why these two words are intertwined.

                                                                        

ML vs AI

Machine Learning is predicting based on the data(experiences) given, whereas the action taken upon is termed as Artificial Intelligence. Humans also do the same thing based on their experiences the action is taken.

                                                                     

ML vs AI

Roadmap for achieving AI:

                                                                       

Steps involved in achieving AI

1. Business Intelligence (BI): It is one of the key factors where in which based on the data, the problem or insights are found and then analysed for the best solution.
2. Big Data: Huge amounts of data is required to do anything substantial in machine learning(mostly deep learning). When storage got cheaper huge amount of data was getting stored. Using Big Data just throw all our data into Hadoop and run batch processes on it called MapReduce that ended up replacing/augmenting our data warehouses. In such a simpler way, the leap from BigData to ML happened mostly because of Deep Learning.
3. Machine Learning: As discussed previously ML will learn things on it's on if enough data and a little guidance are provided. (My next post is about the need of ML and on why explicitly programming is not a good idea). Here certain specific algorithms are used for achieving the expected results.
4. Deep Learning: It is a part of Machine learning where the number of layers involved is increased to achieve better results so that it might outperform all models.
5. Artificial Intelligence: It is the last step involved where we can make the machines smart enough so that it might be on a human level or surpass human capabilities. Here the specific algorithms used are made more as a target by replacing specific algorithms used.