AI Notes PDF

Introduction to Artificial Intelligence LESSON 1: Problem Solving Knowledge Representation and Reasoning Search algorithms in AI are the algorithms that are created to aid in getting the right solution In Artificial Intelligence, search techniques are universal problem-solving methods. Problem-solvi...

Introduction to Artificial Intelligence LESSON 1: Problem Solving Knowledge Representation and Reasoning Search algorithms in AI are the algorithms that are created to aid in getting the right solution In Artificial Intelligence, search techniques are universal problem-solving methods. Problem-solving agents in AI mostly used these search strategies or algorithms to solve a specific problem and provide the best result. Properties of search algorithms to compare the efficiency of these algorithms: Completeness: A search algorithm is said to be complete if it guarantees to return a solution if at least any solution exists for any random input. Optimality: If a solution found for an algorithm is guaranteed to be the best solution (lowest path cost) among all other solutions, then such a solution for is said to be an optimal solution. Time Complexity: Time complexity is a measure of time for an algorithm to complete its task. Space Complexity: It is the maximum storage space required at any point during the search, as the complexity of the problem. Importance of Search Algorithms in AI Solving problems Finding the best solution possible to a given problem/task Search programming Many AI functions can be designed as search oscillations, which thus specify what to look for in formulating the solution of the given problem. Goal-based agents These agents aims to come up with an optimal solution which takes into account all possible factors. Neural network systems In a neural network system, the structure is composed of a hidden layer, an input layer, and an output layer, and nodes that are interconnected. With search algorithms, it would be able to navigate the search space to find the connection weights that will be required in the mapping of inputs to outputs. Types of Search Algorithms Uninformed / Blind Search search tree is searched without any information about the search space It examines each node of the tree until it achieves the goal node. Can be divided into six main types:  1. Breadth-first search 2. Uniform cost search 3. Depth-first search 4. Depth limited search 5. Iterative deepening depth-first search 6. Bidirectional Search Informed Search (Heuristic search) Informed search algorithms use domain knowledge. In an informed search, problem information is available which can guide the search. Informed search strategies can find a solution more efficiently than an uninformed search strategy. A heuristic is a way which might not always be guaranteed for best solutions but guaranteed to find a good solution in reasonable time. Sample algorithms Greedy Search A* Search Knowledge-based Agent in Artificial Intelligence An intelligent agent needs knowledge about the real world for taking decisions and reasoning to act efficiently. Knowledge-based agents are those agents who have the capability of maintaining an internal state of knowledge, reason over that knowledge, update their knowledge after observations and take actions. These agents can represent the world with some formal representation and act intelligently. Knowledge-based agents are composed of two main parts: o Knowledge-base and o Inference system. Knowledge-based Agent in Artificial Intelligence Knowledge base Knowledge-base is required for updating knowledge for an agent to learn with experiences and take action as per the knowledge. Inference system Inference system generates new facts so that an agent can update the KB. An inference system works mainly in two rules which are given as: Forward chaining Backward chaining Knowledge Representation in AI Knowledge Representation in AI is the method of structuring and organizing knowledge so that AI systems can process and utilize it for reasoning and decision-making. refers to the way in which AI systems store, organize, and utilize knowledge to solve complex problems involves the creation of data structures and models that can efficiently capture information about the world, making it accessible and usable by AI algorithms for decision-making, inference, and learning. Relationship between Knowledge and Intelligence Knowledge as a Foundation: Knowledge provides the necessary information, facts, and skills that intelligence uses to solve problems and make decisions. Intelligence as Application: Intelligence is the ability to learn, reason, and adapt, using knowledge to perform tasks and solve complex problems. Interdependence: Knowledge without intelligence is static, while intelligence without knowledge lacks the raw material to function effectively. Synergy: Effective AI systems require a balance of both knowledge (the “what”) and intelligence (the “how”) to operate successfully. Kinds of Knowledge that AI needs to represent Object: An object is a thing or entity that can be identified and described. Event: An event happens at a specific time and place. Performance: Performance is a measure of how well a task is accomplished. Meta-knowledge: Meta-knowledge refers to knowledge about knowledge.  Facts: Facts are statements that are true or false. Types of Knowledge Declarative Knowledge - can be represented as a set of propositions or statements Procedural Knowledge - can be represented as a set of rules or algorithms. Meta-Knowledge – It provides information about other knowledge’s properties, relationships, and context. Heuristic Knowledge – refers to knowledge acquired through trial and error. Structural Knowledge – Structural knowledge is used to create models that describe the relationships between different concepts or entities Knowledge Cycle Approaches of Knowledge Representation Simple Relational Knowledge – set of rules defining the relationships between different objects Inheritable Knowledge – often used to represent hierarchical relationships between objects Inferential Knowledge – often used to represent logical relationships between objects Procedural Knowledge – often used in expert systems or intelligent agents performing tasks or solving problems Applications of Knowledge Representation in AI Expert Systems: These systems use knowledge representation to provide advice or make decisions in specific domains, such as medical diagnosis or financial planning. Natural Language Processing (NLP): Knowledge representation is used to understand and generate human language, enabling applications like chatbots, translation systems, and sentiment analysis. Robotics: Robots use knowledge representation to navigate, interact with environments, and perform tasks autonomously. Semantic Web: The Semantic Web relies on ontologies and other knowledge representation techniques to enable machines to understand and process web content meaningfully. Cognitive Computing: Systems like IBM’s Watson use knowledge representation to process vast amounts of information, reason about it, and provide insights in fields like healthcare and research. LESSON 2: Common Machine Learning Algorithms Linear Regression  Linear regression is a type of supervised learning. Linear regression uses the regression analysis of mathematical statistics to determine the quantitative relationship between two or more variable. Linear regression is a statistical method that is used to predict a continuous dependent variable(target variable) based on one or more independent variables(predictor variables) Benefits of Regression Analysis It indicates the significant relationships between dependent variable and independent variable. It indicates the strength of impact of multiple independent variables on a dependent variable. Types of Linear Regression Simple linear regression: This involves predicting a dependent variable based on a single independent variable. Multiple linear regression: This involves predicting a dependent variable based on multiple independent variables. Difference Between Simple and Multiple Linear Regression: Simple Linear Regression: This is used when there is only one independent variable and one dependent variable(e.g., predicting y from a single x) Multiple Linear Regression: This is used when there are two or more independent variables and one dependent variable. Artificial Intelligence Overview Machine Learning Overview What is Artificial Intelligence? “Artificial” in artificial intelligence (AI) means that it is designed by and is created for humans. AI is a scientific discipline that studies and develops theories, techniques, and application systems that is used to simulate and extend human intelligence.  The term was first coined by John McCarthy in 1956, who defined it as the “science and engineering of making intelligent machines, especially intelligent computer programs”. What is Artificial Intelligence (AI)? Artificial Intelligence (AI) is a term that refers to the capability of machines or computer programs to perform tasks that typically require human intelligence, such as learning, problem solving, and decision-making. What is Machine Learning? Machine learning is a subset of artificial intelligence (AI) that involves the use of algorithms and statistical models to enable machines to learn from data and improve their performance on specific tasks. What is Deep Learning? Deep learning is a subset of machine learning that involves the use of artificial neural networks to enable machines to learn from data and perform complex tasks. AI can be broadly categorized into two types: Narrow or Weak AI o Designed to perform specific tasks or solve specific problems, such as image recognition or language translation.  o Relies of machine learning algorithms that are trained on large datasets to recognize patterns and make predictions based on those patterns General or Strong AI o Aims to replicate human intelligence in a broad range of domains, such as reasoning, perception, and creativity What is Machine Learning? Machine learning is a subset of artificial intelligence (AI) that involves the use of algorithms and statistical models to enable machines to learn from data and improve their performance on specific tasks. Types of Machine Learning Supervised learning o Involves training a model on a labeled dataset, where the correct output is known for each input. o The model then uses this training data to make predictions on new, unseen data. Unsupervised learning o Involves training a model on an unlabeled dataset and allowing it to identify patterns and relationships on its own. o Often used in applications such as clustering, where the goal is to group similar items together Reinforcement learning o Involves training a model to make decisions based on feedback from its environment  o Model learns through trial and error, adjusting its actions based on the rewards or punishments it receives. Advantages of machine learning - Ability to improve over time – as the machine is exposed to more data, it is able to improve its performance on specific tasks known as “learning by experience” - Can be used to make predictions or decisions in real-time, allowing them to be used in applications where speed is critical. Ethical and Societal concerns - Potential bias that may be built into machine learning algorithms. If training data is biased, the model will also be biased, potentially leading to unfair decisions or outcomes. - Potential impact of machine learning on employment, as machines may be able to perform many task that are currently done by humans, leading to significant job losses. What is Deep Learning? Deep learning is a subset of machine learning that involves the use of artificial neural networks to enable machines to learn from data and perform complex tasks. These networks are inspired by the structure of the human brain and are composed of layers of interconnected nodes that process information. Each node in the network performs a simple calculation based on its inputs and outputs a result, which is then passed on to the next layer of nodes. One of the key advantages of deep learning is its ability to perform complex tasks, such as image recognition, speech recognition, and natural language processing. What is Machine Learning? Machine learning is a subset of artificial intelligence (AI) that involves the use of algorithms and statistical models to enable machines to learn from data and improve their performance on specific tasks. Types of Machine Learning Supervised learning o Involves training a model on a labeled dataset, where the correct output is known for each input. o The model then uses this training data to make predictions on new, unseen data. Unsupervised learning o Involves training a model on an unlabeled dataset and allowing it to identify patterns and relationships on its own.  o Often used in applications such as clustering, where the goal is to group similar items together Reinforcement learning o Involves training a model to make decisions based on feedback from its environment o Model learns through trial and error, adjusting its actions based on the rewards or punishments it receives. Advantages of machine learning - Ability to improve over time – as the machine is exposed to more data, it is able to improve its performance on specific tasks known as “learning by experience” - Can be used to make predictions or decisions in real-time, allowing them to be used in applications where speed is critical. Ethical and Societal concerns - Potential bias that may be built into machine learning algorithms. If training data is biased, the model will also be biased, potentially leading to unfair decisions or outcomes. - Potential impact of machine learning on employment, as machines may be able to perform many task that are currently done by humans, leading to significant job losses. What is Deep Learning? Deep learning is a subset of machine learning that involves the use of artificial neural networks to enable machines to learn from data and perform complex tasks. These networks are inspired by the structure of the human brain and are composed of layers of interconnected nodes that process information. Each node in the network performs a simple calculation based on its inputs and outputs a result, which is then passed on to the next layer of nodes. One of the key advantages of deep learning is its ability to perform complex tasks, such as image recognition, speech recognition, and natural language processing.  Supervised learning o Involves training a model on a labeled dataset, where the correct output is known for each input. o The model then uses this training data to make predictions on new, unseen data. Unsupervised learning o Involves training a model on an unlabeled dataset and allowing it to identify patterns and relationships on its own. o Often used in applications such as clustering, where the goal is to group similar items together Reinforcement learning o Involves training a model to make decisions based on feedback from its environment o Model learns through trial and error, adjusting its actions based on the rewards or punishments it receives. LESSON 3: Machine Learning Process Machine Learning Concepts