Artificial Intelligence

artificial-intelligence

What is Artificial Intelligence?

There is a lot of buzz around artificial intelligence at the moment and the term AI seems to be thrown around a lot but what is it exactly to clear things up. Before starting, we have to look at a definition. 

To avoid confusion, we have to go back to the earliest and hence purest definition of AI from the time it was first coined. The official idea and definition of AI was first coined by Jay McCartney in 1955 at the Dartmouth conference. 

Of course, those plenty of research work done on AI by others such as Alan Turing before this but what they were working on was an undefined field before 1955.

McCarthy proposed every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. 

An attempt will be made to find out how to make machines use language from abstractions and concepts, solve the kind of problems now reserved for humans, and improve themselves. 

AI is the machine with the ability to solve problems that are usually done by us with our natural intelligence. A computer would demonstrate a form of intelligence when it learns how to improve itself at solving these problems. 

To elaborate further, the 1955 proposal defines seven areas of AI. Today there are surely more here are described the original seven. 

artificial-intelligence

Areas of AI:

  1. Simulating higher functions of the human brain.
  2. Programming a computer to use general language.
  3. Arranging hypothetical neurons in a manner, enabling them to form concepts.
  4. A way to determine and measure problem complexity.
  5. Self-improvement.
  6. Abstraction: defined as the quality of dealing with ideas rather than events
  7. Randomness and creativity.

After 60 years, it is thought that realistically we have completed the language measure problem, complexity and self-improvement to at least some degree. However, randomness and creativity is just starting to be explored. 

Recently, we have seen a couple of web episodes, scripts, short films, and even a feature-length film co-written or completely written by AI. So, in the definition, you have heard the word intelligence. What is the intelligence? Well, according to Jack Copeland who has written several books on AI.

Factors of Artificial Intelligence:

Some of the most important factors of intelligence are 

  1. Generalization:  Learning that is learning that enables the learner to be able to perform better in the situations not previously encountered.
  2. Reasoning: To reason is to draw conclusions appropriate to the situations in hand. 
  3. Problem solving: Given such and such data and find x.
  4. Perception: Analyzing, ask and environment and analyzing features and relationships between objects. For example, self-driving cars. 
  5. Language understanding: Understanding language by following syntax and other rules similar to a human. 

Strong AI vs weak AI:

Okay, so now we have an understanding of AI and intelligence. To bring it together a bit and solidify the concept in your mind of what AI is, here are a few examples AI machine learning, computer vision, natural language processing, robotics, pattern recognition and knowledge management. There are also different types of artificial intelligence in terms of approach. For example, the strong AI and weak AI.

Strong AI is stimulating the human brain by building systems that think and in the process give us an insight into how the brain works. We are nowhere near the stage yet. Weak AI is a system that behaves like a human but doesn’t give us insight into how the brain works. 

IBM’s deep blue a chess playing AI was an example. It processed millions of moves before it made any actual moves on the chessboard. It doesn’t stop there though, there’s actually a new kind of middle ground between the strong and weak AI. 

This is where a system is inspired by human reasoning but doesn’t have to stick to it. IBM’s Watson there’s an exam like human it reads lot of information, recognize patterns, and builds up evidences to say hey I am X% confident that this is a right solution to the question that you have asked me from the information that I have read. 

Google’s deep learning is similar as it mimics the structure of human brain by using neural networks but doesn’t follow its function exactly. The system uses nodes that act as artificial neurons connecting information. Going a little bit deeper, neural networks are actually a subset of machine learning. 

So what is a machine learning then? Machine learning refers to algorithms that enable software to improve its performance over time as it obtains more data. This is programming by input-output examples rather than just coding. So that this makes more sense? 

Let us use an example, a programmer would have no idea how to program a computer to recognize a dog, but he can create a program with a form of intelligence that can learn to do so. If he gives a program enough image data in the form of dogs and let it process and learn. 

When you give the program, an image of a new dog that’s it never seen before, it would be able to tell that it’s a dog with relative ease. 

Most artificial intelligence algorithms are expert systems. So what are expert systems? The often cited definition of an expert system is as follows:

“An expert system is a system that employs human knowledge in a computer to solve problems that ordinarily human expertise”. 

Basically it’s a practical application of knowledge database. Demis Hassabis was a co-creator of D mind, highlighted in a Google blog that we have mastered go and thus achieved one of the grand challenges of AI.

 However, the most significant aspect of all of this for us is that there are not just expert systems that are built on handcrafted rules, instead it uses general machine learning techniques to figure out for itself. 

Our hope is that one day; they could be extended to help us address some of the society’s toughest and most pressing problems from climate modeling to complex disease analysis. 

In other words, the algorithms that are used to serve as a basis to be applied to very complex problems. 

Applications of artificial intelligence:

A lot of us are paranoid about how artificial intelligence might negatively impact our lives. However, the present picture is thankfully more positive. So, let’s explore how artificial intelligence is helping our planet and at last benefiting humankind.

  1. Artificial intelligence in artificial creativity:

Have you ever wondered what would happen if an artificially intelligent machine tried to create music and art? There is a system which is called muse net which creates music by an artificial intelligence system. 

A MuseNet is a deep neural network that can generate 4 minute musical compositions with 10 different instruments and can combine styles from beetles band. 

MuseNet was not exclusively programmed with an understanding of music but instead, it discovers patterns with harmony, rhythm and style learning on its own. 

Another creative product of artificial intelligence is a content automation tool called wordsmith. Wordsmith is a natural language generation platform that can transform your data into an insightful narrative. Tech giants such as Yahoo, Microsoft are using wordsmith to generate 1.5 billion pieces of content every day.

  1. AI in social media:

Ever since social media has become our identity, we have been generating an unmeasurable amount of data through chats, tweets, posts and so on. Whenever there is an abundance of data, AI and machine learning are always involved. 

In social media platforms like Facebook, AI is used for face verification. Whereas, machine learning and deep learning concepts are used to detect facial features and tag your friends. Deep learning is used to extract every minute detail from an image by using a bunch of deep neural networks. 

Machine learning algorithms are used to design news feed on your interest. Another such examples is Twitter’s AI which have been used to identify hate speeches and terroristic languages in tweets. In makes use of machine learning, deep learning and natural language processing to filter out offensive content. 

According to a recent survey, company has discovered almost 300,000 terrorist accounts 95% of which are found by non-human artificially intelligent machines.

  1. AI in Chatbots:

In these days, virtual assistance have become a very important technology. Almost every household has a virtually assistant that controls the home. A few examples include Siri which is gaining popularity because of the user experience they provide. Amazon Alexa is an example of how AI can be used to translate human language into desirable actions. 

This device uses speech recognition and natural language processing to provide a wide range of tasks on your command. 

It can do more than just to play your favorite songs, it can be used to control the devices at your home, book cabs for you, make phone calls, order your favorite food, check the weather conditions and so on. 

Another example is a google virtual assistant called google duplex that have astonished millions of people. Not only it responds to call and book appointments for you, it adds a human touch. It uses natural language processing and machine learning algorithms to process human language and perform tasks such as manager’s schedule, control your smart home, make reservations and so on.

  1. AI in autonomous vehicles:

For the longer period of time, self-driving cars has been great area of interest in AI industry. The development of autonomous vehicles definitely revoluationalize transportation system. Companies like Waymo conducted several test drives before deploying public ride car services. The artificial intelligence system collects data from the vehicles radar, cameras, GPS and cloud services to produce control signals that operate the vehicle.

 Advanced deep learning algorithms can accurately predict what objects in the vehicles vicinity are likely to do. This makes Waymo cars much more effective and safer.

Another important example of autonomous vehicle is a Tesla’s self-driving cars. AI implements computer vision in which detection and deep learning to build cats that can automatically detect objects and drive around without human intervention. Elion Mask, the founder of Tesla, talks a ton about how AI is implemented in Tesla’s self-driving cars and auto-pilot features.

 He quoted that Tesla will have fully self-driving cars ready by the end of the year and a robot taxi version, that can carry passengers without anyone behind the wheel. 

Tesla’s auto pilot software was beyond driving the car you tell it to go. If you are not in the mood for talking, auto pilot will check your calendar and drop you to your scheduled appointment. This sound pretty amazing!

  1. AI in space exploration:

We have applications of AI in space explorations. This is one of the most interesting fields in which AI is being implemented. Space exploration and discoveries always require analyzing large amount of data. AI and machine learning is the best way to handle and process data. 

After years of research, astronomers use AI to go through years of data obtained by the Hubble telescope in order to identify distant planets and solar systems. This was accomplished by using AI technology.

 AI is also being used in NASA’s rover machine to Mars which was the Mars 2020 rover. AEGIS, which is the NASA’s current Mars rover is already on the red planet. The rover is responsible for autonomous targeting of cameras in order to perform investigation on Mars. This proves that how far has AI reached.

AI Examples

Recent AI inventions:

The current decade has been immensely important for AI inventions. In recent years, AI has become embedded in our day-to-day existence. We use smartphones which have voice assistance and computers that have intelligence functions. AI is no longer a pipe dream. 

In 2010, imagenet launched the imagenet large scale visual recognition challenge also Microsoft launched Kinect for Xbox 360, the first gaming device that tracks human body movement using a 3D camera and infrared detection. 

In 2011, Apple released Siri, a virtual assistant on iOS operating system. Siri depends on natural language user interface to infer, observe, answer and command things to its human user. It adopts the voice commands, projects and individualized experiences for the users. 

Then in 2012, two Google researchers trained a large neural network of 16000 processors to recognize of cats by showing it 10 million unlabeled images from Youtube videos.

In 2014, Microsoft released Cortana, their version of virtual assistance similar to Siri on iOS. Also Amazon created Amazon Alexa, a home assistant that developed into smart speakers that function as personal assistants. 

Talking about the years between 2015-2017, Google deep minds alphago, a computer program that plays the board game co defeated various human champions.

Not just that in 2016, a humanoid robot named Sophia was created by Hansen robotics. She’s the first robot citizen with her ability to see, make facial expressions and communicate. Through AI Sophia has more human like features when compared to other humanoids.

 Finally in 2018, first bi-directional, unsupervised language representation was developed that can be used in a variety of natural language tasks using transfer learning and also Samsung introduced Bigsby of virtual assistant. Its function includes voice, where the user can speak to and ask questions, recommendations and suggestions.

Limitations of AI:

AI can beat humans at board games such as checkers, chess and go but it doesn’t understand the concept of a game or even the purpose of playing games. Furthermore, an AI which has become a champion of chess could not be used to learn how to play a musical instrument. We will have to need a separate and completely different architecture for that task. 

Narrow AI can make recommendations for movies and purchases on the basis of what you have watched or purchased in the past but it has no idea what it means to watch a movie or buy something. 

It can’t even explain or articulate why it makes recommendations it does. Narrow AI gather information around the globe for use in making weather predictions but it has no idea what weather actually is. 

It has no real world experience. In short, narrow AI works within a very limited context and cant solve problems outside of its training. 

For example, an AI that is designed to detect abnormalities in x-rays cant in turn around and drive you home after work. Even applications like self-driving cars often use a whole range of machine learning. Algorithms to complete that task. 

Narrow AI is very good at processing mountains of data and finding patterns and correlations but it lacks any real common-sense understanding of what that data means or predictions it makes. AI systems often use an approach to processing data and solving problems. Its very different to the way human intelligence does.

For example, if two images are introduced with a little bit of voice inside it. For humans, it will be a no problem for identifying image before and after noise but for narrow AI it will require a complex system.