INTRODUCTION:
All you need to know about AI or Artificial intelligence is the intelligence processed by machines designed by humans. It can work like a human almost anything to which it is designed. John McCarthy was the father of Artificial intelligence. Machines learn from experiences, more the data inputs in it, the more will be the efficiency.
EXAMPLE:
It can be understood easily by this example. Let’s say you are using words, sheets, or anything like that, and you are making dozens of sheets simply, All you need to know about AI, you are giving so much input, but it can not increase its efficiency. It will be the same as it was earlier. On the other hand, if a camera is installed on the road working with the help of AI, with every passing car it will be improved and its detection efficiency will also improve.
LIMITATIONS:
Although AI is highly capable, it has limitations like whether it is designed to work like humans but it can’t to a greater extent, it doesn’t have a human-like understanding, it depends on the data, insufficient results in flawed outcomes, limited creativity, reliance on computing power, unethical privacy concerns and of course lack of common sense.
TYPES OF AI: (All you need to know about AI)
AI is classified on the terms of:
Based on Capabilities:
- Narrow AI
- General AI
- Super AI
Found on Functionality:
- Reactive machines
- Self-awareness
- Theory of Mind
- Limited theory
BASED ON CAPABILITIES:
NARROW AI (ANI):
Narrow AI is also known as weak AI. It is a type of artificial intelligence in which an algorithm is designed to perform a single task. So, it is application-limited. Programmers provide specific domain models. It learns from thousands of labeled data (examples).
Furthermore, knowledge of one domain cannot transfer to another. It is today’s AI.
EXAMPLE:
Virtual assistants like Alexa, Siri, chatbots, face and image recognition etc work on Narrow AI.
GENERAL AI (AGI):
General AI is the AI that is able to perform human-like tasks. It refers to the broader AI system that can perform intellectual tasks that a human mind can do using its own mind. But it is really challenging and still research is ongoing.
EXAMPLE:
Sophia robot was the first AI robot and somehow works on general AI. However, research is still ongoing. But it can be our future AI.
Super AI (ASI):
It is a hypothetical AI system. In this, AI machines can perform tasks more intellectually than humans. It can solve problems that humans can’t.
BASED ON FUNCTIONALITY:
Reactive Machines:
As the name suggests, it reacts to the data or specific (provided) tasks and does not have capabilities beyond these. But it works only in the present or simply says works with present data. Past data can not be stored. The past data are not able to determine future actions. For example, simple status machines and the most famous IBM deep blue ( a chess-playing supercomputer that works on AI as reactive machines ), defeated the grandmaster because this IBM understands all the rules recognizes the chess board, and knows all the moves.
Limited Memory:
Limited Memory AI is exactly the opposite of the reactive machines AI. This AI works from past data and trains itself according to it. But these experiences are short-lived and may not live long.
For example, this technology is used in traffic lights and self-driving vehicles. Limited Memory AI recognizes other vehicles and how they are moving around, changing lanes, etc.
Theory of Mind:
Theory of Mind AI exists as a concept or hypothetical. It is the top advanced class of technology. This AI is designed to understand the emotions and behaviors of humans and their interaction with the environment. So the theory of mind AI needs a thorough understanding of it.
For example,
One popular example of the real theory of mind AI is Kismet. It was invented in the late 90s by a Massachusetts Institute of Technology researcher. Kismet can mimic human emotions and recognize but cannot convey attention to humans.
Another popular example is the Sophia robot. It was invented by Hanson Robotics. Theory of Mind AI is implemented in this robot and the camera is present in Sophia’s eyes. She was able to recognize individuals, sustain eye contact, and also can follow faces.
Self-awareness:
This is the hypothetical type of AI. This self-awareness AI will be smarter than the human mind. This system can understand their internal traits, and conditions and also perceive emotions. And they also have their own emotions, beliefs, and needs on their own. They will have these 4 types of things:
- Emotional intelligence ( recognize and manage their own emotions)
- Consciousness (unified sense and awareness of existence)
- Self-assessment (evaluate and identify own performances for improvement)
- Reflective learning (learning from its own mistakes)
CONCLUSION:
AI is a vast branch. Day by day we are making progress in AI and in the upcoming years we will be able to see machines on general AI hopefully.
Today, we are far from self-aware and creating machines that can solve all the problems but we should make our efforts to understand thoroughly on this branch.
All you need to know about AI
