Why the industry 4.0 is hell bent on using AI/ML?

So, before we begin what Artificial intelligence is, I would like to tell you a few snippets of news that make us wonder, wow can this …actually…be done? Damn… son!!

The Most used app this lock down, Netflix, Duh, I’m sure it needs no introduction. It is a extensively popular video content delivering service that uses analytically predicting knowledge to propose suggestions on the basis of how consumers’ react, what is their interests, what were their previous choices, and currently their series or movie watching behaviour. This is in need of examines from several data as said above to recommend movies based on your earlier liking and watch history.

With Each Passing day, Netflix is turning more intelligent. From creating content as the user wants and to recommending the content the user might like. With this mentioned, the app is also working on increasing the growth of small movies that would go unnoticed in the usual setup of movie culture, while bigger films grow and propagate the theatres. To have this as a full-time intelligent software will take years to come but nevertheless this is a practical example that shows how AI is growing.

if you are still not convinced, below are few stats that was published on the internet, this will give you an idea of growth of AI In the markets!

– In 2014, more than $300 million was invested in start-up’s, showing an increase of 300%, compared to the previous year (As Said by Bloomberg)

–  Artificial intelligence will replace 16% of American jobs by the end of the decade (By Forrester in early 2K15’s), This was done by a vastly implementation of Robotic Process Automation and did create a huge amount of opportunities in the market rather than job cuts.

– By 2023, the number of people with disabilities employed will triple due to AI and emerging technologies reducing barriers to access. (As said by Gartner)

– As of today, over 55% of Apple phone owners’ users use Siri’s voice recognition capabilities, which is an iconic example of the Machine learning technology in devices.

Unlike overall perception, artificial intelligence is not limited to just Information Technology or Tech industry; instead, it is being widely used in other areas such as medical, business, education, law, fashion and manufacturing.We are writing down what AI is and what are the other intelligent AI/ML products that we are using, promoting the interest to learn machine learning and AI in you.

So, what is AI? Who found it? Why?

As we study in school and colleges nowadays, lets define AI! Shall we..

Artificial intelligence is an intelligence demonstrated by any machine, which are derived as a substitute in some cases or an addition to the raw intelligence of the man such as learning and solving issues based on categorizing it into problems and priorities.

Sofia – The Robot

Well, if it’s up to me I would say Alan Turing is the original father of the machine learning interfaces after we saw the movie the Imitations game, where he let the machine handle the outcome of a battle favouring Britain with fewer deaths, the ENIGMA, German intelligence’s finest crypt messaging device was broken, Christopher for the win. Damn that happened… It is a true story!!!

Anyhow… I am Not!

So let’s see what the books and events in our history define, they say John McCarthy to be the real father of AI!!  Why? Coz he was a smart person, and more so, because coined the term “Artificial Intelligence” in a Dartmouth conference and then went on to invent an programming language names LISP, the oldest programming language after FORTRAN obviously. After LISP was beaconed out to the world researchers and scientists and other people were forced into solving problem using the computers and there was a surge in inventions of newer algorithms. With the emphasis on the problem solving problems and creating robots with learning abilities, WABOT-1, a intelligent humanoid was built in 1970’s

WABOT 1

So, cut to 2020, today, the advancements have made it possible to design many such intelligent robots namely-> ASIMO, Atlas, Unimate, AIBO and Sophia. The latest Robot Machines that exhibit the consciousness and activities that humans can perform and obviously more.  

Asimo
Atlas
Unimate – the first industrial robot
AIBO – The Cutest dog that never leaves your side

Well, That’s AI plus the advanced version of Machine learning, a combo, that let these sophistication in the technology possible.

Let me also mention the chess expert robot, in 1997, IBM’s ‘Deep Blue’ defeated Garry Kasparov and it became the first computer to beat a world chess champion.

Deep Blue Robo and its handler trying to piss off the champ and being successful 😀

Now, A vast increase in computer processing power and storage abilities have now allowed companies to store huge quantities of data through servers and clouds. In the past 5 years, Amazon, Google, Tesla and other companies have leveraged machine learning to their huge commercial advantage and advancements in the technological revolution. The best part to be adored and respected is that these companies do not just process user’s data to understand consumer behaviour for their profit, these companies have continued the work on vision, natural language processing, and a lot of other AI applications in various domains. As a result, today, the technology sector drives the stock market in most countries.

So, the main question still remains to be answered why AI for all industries is needed?

Let’s say many organizations and tech companies are investing money on data as a source of future revenue, the same data has to be ordered and extracted to make meaningful information which is performed by the data scientists with domain a strong domain knowledge, simply put they know what they are looking for in the gibberish(for more info on data science visit the following link: https://arhaaths.co.in/2020/01/30/331/).

So Now, The extraction through human known entities are limited to a few and companies cannot afford to employ 1000’s of data scientists with expertise in various domains. And one candidate cannot possibly know all about many domains at once. He is bound to show human characteristics and be confused. Hence the AI assist here, algorithms that improve themselves as they encounter data can be priceless. So we as AI Engineers and ML Engineers write supervised algorithms where we give the machine a few hints and ask it to find out what we intend to find int the first place. This is called training in the fields for a ML Algorithm.

So how does that training work? 😛 Six packs much?  

Supervised machine learning algorithms are ‘trained’ on data whose fields have been ‘labelled’ like needs vs want for humans.

Example, In a graph (X,Y) is labelled, say X is the number of cars sold & Y is the type of a car.

 Then, they sort new data plus incoming data according to their training. 

However, collecting several input-output paired samples is high-priced in various practical applications of everyday, so we have other techniques that can help us out in this.

Other kinds of Learning algorithms include unsupervised learning, semi-supervised learning, and reinforcement learning.

However, what truly characterizes machine learning algorithms/AI to be used in the idutry is that it is able to cope with variety of complex tasks that a normal human being would struggle to work out.

But having said that there is no General AI as of today because, even the greatest notable ML application, the DeepMind’s AlphaGo computer program are focused only on a narrow task or set of tasks. There is no evidence of a AI/Robot or an ML that can do all events and attain the general AI mind.

These industries are on a race towards to finding this greatest invention of all time, Artificial General Intelligence or as some texts call it “STRONG AI”. The significant progress in artificial intelligence in the last few years as mentioned above, indicates that a strong artificial intelligence may be attainable in the near future.

And like how I always say when people ask me WHY? -> Because Why NOT?

Having said all these, here’s a few applications of AI that we think are possible in some domains:

Also, We are in an era where the growth of services like AIaaS are close to blasting out of proportion,

Yes, You saw that perfectly all right!!!

Artificial Intelligence as a Service is the way to get a third party hired, who are offering AI outsourcing. This service allows people and enterprises to conduct experiment with AI for various objectives without heavy financial investments and also get away with a lower risk in case the experiment fails. Experimentation here allows the sample of different data platforms to test every machine learning algorithm’s efficacy for the creation of newer revenue streams.

A System that is used for solving problems must surely have to consider and require human skill like it should be able to learn, plan, reason right or wrong based on situations, perceive complex events, and then get down to solving problems.

Lets take Healthcare as the first domain under the AI knife,

AI and IoT/ML can be merged into developing applications that help improve tools for scanning patients and reduce costs for stuff like x-ray/MRI etc..,. Let’s say Corona detection as done by an IIT professor, he mixed and matched Machine Visions and detection of CORONA patterns in lungs of patients far better than the available x-ray machines and even beat the doctors who are equipped to handle this virus. AI can also be used to build chatbots that read symptoms and then provide medical answers on the internet, or analyse the condition of the patient, order ambulance for pickups, alert specialist doctors who are nearby, and schedule doctor appointments.

AI as an education and its uses in education,

By observing the trend of how many IT employees are trying to make a breakthrough learning AI, the education systems needs to realise the importance of spreading the AI /ML literacy to all schools and bachelor colleges so that students are ready to make a difference in the world outside.

Say a practical application for students can be an AI that determines how they best perform in exams based on scores and determine a pattern of learning for the coming years. It can shape a whole new stream of schooling where we enable virtual tutors for kids that need special attentions and also cater the over smart top of the class students to their level of higher challenges to make them extra ordinary. This types of AI can be extended to make the students stay confident even if they fail as human bias Is removed and the true analysis lets student chose his own pathway to success in life rather than a forced ENGINEER or DOCTOR or a LAW degree..

Whooo, that was nerve racking right? More to come… 😀

Let’s say corporate you work in, needs AI to be instilled in all its existing processes. As we know consumer preferences are changing by the minute. AI, after processing enough data about consumer, can help us to write of the predictions as to what the user might want and help the businesses stay relevant to the client and keep them happy with the services provided. Chatbot’s and RPA’s are the kind of AI that will speed up things and help the business magnets stay magnets 😛

Here again, Hiring can be made simpler and less biased, clean and only based on true skills.

PS: I AM NOT SAYING THINGS ARE CORRUPT but if you know you know. NO CONTROVERSIES HERE PLEASE. 😛

The real difference of AI and robots will be felt when AI-powered robots will be replacing the human workforce, Yes I watched the I-ROBOT movie 😊. This can be a debatable topic for debates, but to see the positives side, This could lead in reduction of humans down the drains and jobs that require utmost risk where humans are at risk. This could make houses, streets, cities cleaner with bots with invention of cleaner bots, hence stop spreading bacteria, fungi, viruses etc..,

AI could also help tailor creative solutions to security problems like theft and impersonations etc., AI Bots can help in research of finding the multiverses, answering some existential questions about aliens, which will lay the foundations to understanding the universe better.

This I hope, will lead to a newer paradigm in our lives, where it is are made easier to live a full life and the value of time and money are realised and create a world where no one is judged and we all get to live in real harmony.

Few tutorials on AI that will get you started are: (Not ranked)

1.) Elements of AI: they are really good in demystifying the concepts and help you understand better.

Link: https://www.elementsofai.com/

2.) What’s better than getting a IBM cert for AI and kickstarting your AI career with WATSON-AI by understanding hoe real world AI apps work and how you can build one:

https://www.coursera.org/professional-certificates/applied-artifical-intelligence-ibm-watson-ai

3.) Become a master with Arizona State University on AI & ML:

https://www.coursera.org/mastertrack/ai-machine-learning-asu

4.)EDX courses (from Harvard only)

https://www.edx.org/course/cs50s-introduction-to-artificial-intelligence-with-python

https://www.edx.org/course/data-science-machine-learning

5.)ANDREW NG’s ML course : https://www.coursera.org/learn/machine-learning

6.)Learn about AI/ML From experts at GOOOGLE: https://ai.google/education/

7.) Artificial Intelligence A-Z™: Learn How To Build An AI(UDEMY) : https://www.udemy.com/course/artificial-intelligence-az/

If you don’t like any of them and want more options, https://www.classcentral.com/subject/ai. This page will definitely help you.

Please let me know if this article has interested in tickling you AI/ML senses.

Note: The images taken are from Google and are for representational purpose!

Advertisements

2 thoughts on “Why the industry 4.0 is hell bent on using AI/ML?

Leave a Reply

%d bloggers like this: