Edcorner Learning, Edcredibly Team. Artificial Intelligence - Logic & Algorithms for problem solving. Volume 2

Independently published, 2021. — 701 p. — ASIN B096KYY5RX.Artificial intelligence is, at its core, a system that can perform a task using intelligence that mirrors (or is better than) human intelligence. Theoretically, any task that requires human intelligence to accomplish could instead be performed by artificial intelligence assuming the system has the adequate information and capabilities programmed. It accomplishes this by utilizing processes such as machine learning to scour sets of data and utilizing algorithms (instructions, or list of rules a computer should follow to solve a problem), to discover trends in data and provide insights for decision-making. This book will give you practical knowledge about different logics and algorithm to more than 140+ Problems than can be solved by AIAI Module 1 Introduction
The problems that need special attention
Why study AI
AI techniques
Heuristic based search
Knowledge representation and inference
Reason with incomplete information
Fault tolerance
AI Module 2 State Space Search IState Space
Solving AI Problems
Examples of production rules
Applying rules to solve the problem
AI Module 3 State Space Search II
A state space search for a problem with more prerequisite
A missionary cannibal problem
The farmer fox chicken grain problem
The combinatorial explosion
AI module 4 Guided and unguided searchGuided and unguided search
Generate and test
Breadth first search (BFS)
Depth first search (DFS)
Depth bounded DFS (DBDFS)
Comparison
AI Module 5 Heuristic search methodsHeuristic function
Hill climbing
Best first search
Branching factor
Solution space search
AI module 6 Other Search methodsVariable neighbourhood decent
Beam search
Tabu search
Simulated annealing
AI module 7 Problems with search methods and solutionsLocal and global heuristic functions
Plateau and ridge
Frame problem
Problem decomposability and dependency
Independent or Assisted Search
Search for Explanation
Iterative Hill Climbing
AI Module 8 Genetic algorithm & Travelling salesman problem
Genetic Algorithms
Basic operations
Selection
Recombination
Mutation
Traveling salesman problem
PMX (Partially Mapped Crossover)
OX (Order Crossover)
CC (Cyclic Crossover)
Other RepresentationsAI module 9 Neural networksBrain and CPU works differently
The artificial neural networks (ANNs)
The Neuron and the ANN
The process of learning
Learning for correct values and speed of learning
Generalization
The black box of reasoning
Unsupervised Learning
AI module 10 Multi-layer feed forward networks and learning
Prerequisites to the Backpropagation algorithm
Choosing number of nodes at each layer
AI module 11 Learning in Back Propagation networkLearning in Back Propagation network
Geometrical view of the learning process
Content addressability and Hopfield networksAI module 12 Ant Colony Optimization, branch and bound, refinement search.Ant Colony Optimization
How ants discover optimal paths
Solving TSP problem using ACO
Calculating the pheromone value
Branch and bound
AI module 13 The A* AlgorithmPrerequisites for A*
The Graph Exploration using A*
A* algorithm version 1
What if the g value is not identical for the entire path? How
paths are explored
The A* algorithm version 2
The back propagation of estimates
Why h’ and not h?
AI Module 14 Admissibility of A*, Agendas and AND-OR graphsAdmissibility
The effect of g
The effect of h’
Agenda Driven Search
The AND-OR graphs
AI Module 15 Iterative deepening A*, Recursive Best First, AgentsIDA*
IDA* algorithm
Limitations of IDA*
Recursive best first search
Agents
Agent Environment
Rationality
Learning
AI Module 16 Introduction
Objectives and planning
Types of planning
Agent Based Planning
Forward planning
Backward planning
Choosing between forward and backward reasoning
AI Module 17 ProgressionProgression
Relevant and non-relevant actions
Regression for goal directed reasoning
Goal Stack Planning (GSP)
GSP example
Testing the validity of a planAI Module 18 Problem with GSPProblem with GSP
Sussman’s Anomaly
Another route
Plan space planning
Solving Sussman’s anomalyAI module 19 Game Playing AlgorithmsCharacteristics of game playing algorithms
History
Types of Games
Game treesAI module 20 Prerequisites to MiniMax and othe ralgorithmsThe process of MiniMax
Static Evaluation FunctionAI module 21 MiniMax algorithmFunctioning of MiniMax Algorithm
MiniMax Algorithm
The process
Need for improvementAI Module 22 Alpha Beta cutoffsMiniMax with Alpha Beta Pruning
Algorithm
The process
Futility cutoffAI Module 23 Other RefinementsWaiting for stability
Look Beyond the Horizon
Using predetermined moves
Use other algorithms
State Space Search *(SSS*)
B* searchAI Module 24 Propositional and Predicate logicFormal Logic
Entailment in Formal Logic
Proportional logic
Need for Predicate logic
Predicate Structure
Using Universal and Existential quantifiers
Representing facts and rulesAI Module 25 Using Predicate logicThe impact of universal and existential quantifiers
Incomplete information
Answering a question
Using functions
Rules that do not work
Unification processAI Module 26 ResolutionConversion to Clausal form
Producing a proof
Proving using resolutionAI Module 27 Knowledge representation using NMR Sand ProbabilityProblems with predicate logic
Non-monotonic Reasoning system
The basis for non-monotonic reasoning
NMRS Processing
Uncertainty and related issues
Statistical reasoning and Probability
Bay’s formula
Certainty factorsAI Module 28 Using Fuzzy logic, Frames and Semantic Net for knowledge representationThe need for Fuzzy logic
Fuzzy sets and fuzzy logic
Using multiple Fuzzy Sets to implement rule
Frames
Frame Systems
Semantic Networks
The importance of indicating objects
Representing quantification
AI Module 29 Stronger knowledge representation methods: Conceptual DependencyConceptual Dependency
Primitives actions for CD
Conceptual categories
Conceptual Roles and Tenses
Syntactical RulesAI Module 30 Syntactical rules for CD and CD’sSyntax rules
Using fuzzy names
Some complex cases
Advantages and Shortcomings of CD
AI Module 31 ScriptsScripts
Some other similar attemptsAI Module 32 Introduction to Expert SystemsES Tasks
What ES entails
The ES Problem Solving
Two different types of ES knowledge
Types of domain knowledgeAI Module 33 ES architecture and Knowledge EngineeringES Architecture
Query processor and client modelling
Interface
Knowledge storage and maintenance
Knowledge Engineering
The inference logic
Updating Knowledge
Explanation system
ES levelsAI Module 34 ES Development process-ISE challenges
ES Development steps
Identification
Identifying the problem
Assessment of applicability
Availability of the expert
Defining the scope
Economic feasibility
Final SelectionAI Module 35 ES Development process-IIPrototype Construction and Conceptualization
Formalization
Project planning
Test Planning
Product release planning
Support planning
Implementation Planning
Implementation
Testing and Evaluation
Performance assessmentAI Module 36 Machine LearningMachine Learning
The process of learning
The ingredients of machine learning process
Supervised and Unsupervised learning
Training testing and generalization