Reading Time: 2 minutes

Artificial Intelligence has become one of the fascinating technology among futurologists and scientists. The capability of an AI machine is limitless. But more than that is the likeness of the AI machine with the human being.

Artificial intelligence has been classified into three main categories:-

1. Machine learning is used for designing new learning algorithms and enabling the computer with the improved existing algorithms also. The technique is used for various purposes such as pattern recognition to make predictions or adjustments.

 Machine learning of three types:- Supervised, Unsupervised, and reinforcement learning.

2. Robotics are used for interacting with people and the world predictably. Deep learning techniques in robots are also used for situation manipulation and other purposes. Robotics can be divided into- soft robotics, swarm robotics, touch robotics, humanoid robotics, and serpentine robotics.

3. Artificial neural networks (ANN) are focused on algorithms development to mimic the human brain functioning, mainly, neocortex area. However, the technique is not considered correct, since the neurons in the brains are not in a linear sequence, used for ANNs.

The three different types of ANNs are Deep learning, convolutional neural networks, and recurrent neural network.

The AI is said to have evolved into four phases:-

1. Phase 1 is known as Reactive machines.
   • AI systems are basic.
   • Past experiences cannot be used for current decisions to be influenced.
   • Deviation from the assigned tasks is not possible.
   • Interactive participation is not possible with the world.
   • A particular situation is handled in the same way each time.

Example:- IBM’s Deep Blue.

2. Phase 2 is known as Memory-forming machines.

• Memories can be formed from past experiences.
• The significant improvisations cannot be made from the memories.

Example:- Self-driving cars.

3. Phase 3 is known as Partially aware machines.

• The current technologies with limited success are trying to reach this stage.
• The other entities can influence the AI systems to learn contextually.

Example:- Video simulations are used by the US Navy for robots to learn human ethics.

4. Phase 4 is known as Fully aware machines.

• AI evolution final evolution.
• Self-aware and self-conscious AI.
• AI learning from scratch without instructions.

Example:- Deep Q network creators can play video games without instructions.

Source:- Statista