Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL)

1. Artificial Intelligence (AI) Fundamentals

• Definition: Artificial Intelligence (AI) is a field of computer science focused on building systems that can perform tasks that normally require human intelligence. These tasks can include understanding language, recognizing patterns, making predictions, solving problems and making decisions.
• AI does not necessarily mean that a machine thinks exactly like a human. Many AI systems are designed to perform specific tasks using algorithms, data and computational models.
• Purpose:Making a Machine intelligent to work like human.
• Founder/Father of AI is John McCarthy .
  • AI was formally initiated in 1956
  • Conference: The term was first coined in 1956 at the Dartmouth Conference.
• First Demonstration: The first demonstration of an AI program ran at Carnegie Mellon University.

Major Areas Related to Artificial Intelligence

AI is a broad field that includes several areas of study and application. Common areas include:

1. Machine Learning (ML)
2. Deep Learning (DL)
3. Natural Language Processing (NLP)
4. Robotics
5. Expert Systems
6. Fuzzy Logic

There are 4 Types of AI:

1. Reactive Machines AI
2. Limited Memory AI
3. Theory Of Mind AI
4. Self-aware AI

AI Programming Languages::

• Common Languages use to build AI is Python, Lisp (List Processing), Java, JavaScript, C++, Prolog (Programming in Logic), Julia, Haskell, Scala, and R.
• Python is one of the most widely used programming languages for AI and machine learning because it has a large ecosystem of libraries and frameworks such as NumPy, pandas, scikit-learn, PyTorch and TensorFlow.
• LISP (List Processing) was created by John McCarthy (father of AI).
• Prolog, R, Julia are also considered as AI Programming Language.

3. Machine Learning (ML)

• Machine Learning (ML) is a branch of Artificial Intelligence in which algorithms learn patterns from data and use those patterns to make predictions, classifications or decisions. The model's performance can improve when it is trained with suitable data and evaluated properly.
• Machine Learning focuses on the use of data and algorithm. Data Mining is the process of discovering useful patterns, relationships or trends in large datasets. Machine-learning techniques can be used as part of data-mining workflows, but data mining and machine learning are not the same thing.
• Founders:
  • Term 'Machine Learning' coined in 1959 by Arthur Samuel (IBM).
  • There isn't a universally accepted single "father/founder" of machine learning. Important contributor: Geoffrey Hinton is one of the major pioneers of neural networks and modern deep learning.
• Data Mining Connection: Application of machine learning methods to large databases is called Data Mining.

4. Types of Machine Learning (ML)

There are 3 Types of Machine Learning:

• 1. Supervised Learning:
  • Supervised learning uses labelled training data, where the expected output is known. The model learns the relationship between the input and the known output.
• 2. Unsupervised Learning: Unsupervised learning works with data that does not have labelled outputs. The algorithm attempts to discover patterns, groups or structures in the data.
  • Requires no supervision
  • Objective: Discover hidden patterns or data groupings without human intervention.
  • This is used for finding hidden structure in unlabeled data.
  • Neural Networks, Anomaly detection.
• Reinforcement Learning (RL)
  • Works on a feedback-based process using the Hitting & Trail method.
  • It creates an autonomous, self-teaching system.
  • Applications include Real-Time decisions, Game AI, and Robot Navigation.

5. Deep Learning (DL)

• Deep Learning is a subset of Machine Learning that uses neural networks with multiple layers to learn complex patterns from data. Deep-learning models can be trained on different types of data, including images, audio, text and numerical data.It teaches computers to process data in a way inspired by the human brain.
• Deep learning models are neural networks with three or more layers.
  • Input layer: Data enters.
  • Hidden layers: Process and transport data.
  • Output layer: Final result or prediction is made.
• Examples: Self-driving cars, Facial recognition, Medical science, Speech recognition, Digital assistants, Fraud detection.

6. Key AI Languages, Robotics, and Chatbots

Programming Languages for AI:

• Python (often considered the best choice), LISP (LISt Processing - historically significant and the first AI programming language called IPL), PROLOG (Programming in Logic), Java, JavaScript, C++, Julia, Haskell, Scala, R.
• LISP Creator: John McCarthy.
• Not Commonly Used (Tested): Perl.

Robotics History:

• Robot Arm: A robot’s "Arm" is called a Manipulator (also Actuator/Effector are related parts).
• First Robot: Introduced in 1960 to the General Motor’s Assembly Line.

Chatbots and Virtual Assistants Chronology:

Product / Chatbot	Developer / Company	Year/Details
First Chatbot: Eliza	N/A	Introduced in 1961.
Siri (Virtual Assistant)	Apple	Year: 2011.
Alexa (Virtual Assistant)	Amazon	Year: 2014.
Cortana (Virtual Assistant)	Microsoft	Year: 2014 (End Services: March 31, 2021).
Bixby (Virtual Assistant)	Samsung	Year: 2017.
ChatGPT (AI Chatbot)	OpenAI (San Francisco startup)	Initially released November 30, 2022. The acronym GPT stands for Generative Pre-trained Transformer. Trained using RLHF (Reinforcement Learning from Human Feedback). ChatGPT is a conversational AI system developed by OpenAI and was publicly launched in November 2022. It uses generative AI models to produce responses to user prompts. It is a Large Language Model (LLM).
Bard / Gemini	Google	Rivals ChatGPT.