Adversarial Problems

Adversarial Problems

• start the course
• describe adversarial problems and the challenges they impose on AI
• specify how to represent an adversarial problem
• describe how to use the minimax algorithm to play an adversarial game and some of its shortcomings
• describe how to use alpha-beta pruning to improve the performance of the minimax algorithm
• describe evaluation functions
• describe how to use cutoffs to be able to perform adversarial searches under a time constraint
• describe how lookup tables can be used to improve an agent's performance
• describe chess and how agents can be made to play the game of chess
• describe expectiminimax values in stochastic games and how they make solution searching harder
• describe different evaluation functions that can be used to search in a stochastic game
• describe how to use monte carlo simulations to make decisions when searching
• build a full high-level representation and solution for an adversarial game using the minimax algorithm and alpha-beta pruning

Constraint Satisfaction Problems

Constraint Satisfaction Problems

• start the course
• define constraint satisfaction problems and describe how they are different from search problems
• list some examples of problems that are better for constraint satisfaction algorithms than search algorithms
• describe how to use a backtracking search to solve a constraint satisfaction problem
• describe how to order variables when performing a backtracking search
• describe arc consistency and other types of constraint consistency in a constraint satisfaction problem
• describe how to use arc consistency to solve a constraint satisfaction problem with constraint propagation
• describe how to use the backjumping and forward checking inference method in a backtracking search
• describe how local search algorithms can be used to solve constraint satisfaction problems
• describe how to represent a Sudoku puzzle and how to solve it as a constraint satisfaction problem
• build a full high-level representation and solution for a constraint satisfaction problem

Introducing Natural Language Processing

Introducing Natural Language Processing

• start the course
• define NLP, and list some of its applications and methods
• describe some of the base NLP operations such as Regex, tokenization, and stemming
• describe the porter stemming algorithm used to stem English text
• describe named entity recognition and some of the methods used to perform this task
• describe how NLP models are built and the various parts required to build them
• describe text classification, why it's useful, and how to perform it
• describe the Naïve Bayes classification algorithm and how it can be used as a simple text classification algorithm
• describe information retrieval and some of the base techniques used to perform this task
• describe how an AI agent can use simple information retrieval techniques to answer some simple questions
• describe parsing and how it can be accomplished using NLP
• describe the challenges related to machine translation and some of the methods used to accomplish this task
• describe methods used by computers to recognize speech
• describe many different types of operations that can be used with natural language processing

Introduction to Artificial Intelligence

Introduction to Artificial Intelligence

• start the course
• describe the four main definitions of artificial intelligence
• describe some of the fields in artificial intelligence research and their applications
• list some of the techniques used to build artificial intelligence systems
• define intelligent agents
• describe the different types of intelligent agents
• define the task environment that intelligent agents live in
• distinguish between an observable, a partially observable, and an unobservable environment, and describe how these affect agents
• describe how the number of agents in a given environment can affect an agent
• define deterministic and stochastic environments and how the level of certainty in an environment affects agents
• describe the different types of environmental behavior and how this can affect agents
• create a description of an environment related to a particular problem and how an agent might behave in that environment

Machine Learning

Machine Learning

• start the course
• describe how AI learns and the different types of machine learning
• describe how examples can be used for learning
• describe decision trees and how the model expresses knowledge
• describe entropy and information gain for learning decision tree models
• describe how to choose attributes to learn a decision tree
• describe overfitting and how decision tree models can be made to mitigate this issue
• describe neural networks and how they apply to artificial intelligence
• describe the structure of a neural network and its individual neurons
• list some of the common types of neural networks and what problems they might be good at solving
• describe how machine learning works with a perceptron
• describe how perceptron learning can be generalized to a multilayered neural network
• describe convolutional neural networks
• describe recurrent neural networks
• describe how a perceptron can learn how to achieve a particular result given a set of examples

Reinforcement Learning

Reinforcement Learning

• start the course
• describe reinforcement learning and list some of the techniques that agents can use to learn
• describe additive rewards and discounted rewards
• describe passive learning
• describe how to use direct utility estimation for passive learning and how to define the Bellman Equation in the context of reinforced learning
• describe temporal difference learning and contrast it with direct utility estimation
• describe active learning and contrast it with passive learning
• describe exploration and exploitation in the context of active reinforced learning and describe some of the exploration policies used in learning algorithms
• define Q-learning for reinforced learning
• describe the different parts used in Q-learning and how these can be implemented
• describe on-policy and off-policy learning and the difference between the two
• describe why lookup tables aren't ideal for most reinforced learning tasks and how to build some function approximations that can make these problems possible
• describe how deep neural networks can be used to approximate q-value for given states in Q-learning
• describe Q-learning and how to set up the algorithm for a particular problem

Search Problems

Search Problems

• start the course
• define search problems and how these can be used by AI agents
• list some problems that are ideal for searching algorithms
• define how to represent search problems
• describe the breadth-first search algorithm
• describe the depth-first search algorithm
• describe depth-limited search and the iterative deepening search algorithms
• describe the greedy approach for best-first informed searching
• define heuristics and their various properties
• describe how to create a good heuristic function for a given search problem
• describe the A* search algorithm
• describe local searching and the hill-climbing search algorithm
• describe the simulated annealing search algorithm and how it improves on hill-climbing search
• describe the three environmental characteristicsof search problems, state the function for a consistent heuristic, and state the function for an A* search

Uncertainty

Uncertainty

• start the course
• describe uncertainty and how it applies to AI
• describe how probability theory is used to represent knowledge to help an intelligent make decisions
• describe utility theory and how an agent can calculate expected utility of decisions
• describe how preferences are involved in decision making and how the same problem can have different utility functions with different agents
• describe how risks are taken into consideration when calculating utility and how attitude for risks can change the utility function
• describe the utility of information gain and how information gain can influence decisions
• define Markov chains
• define the Markov Decision Process and how it applies to AI
• describe the value iteration algorithm to decide on an optimal policy for a Markov Decision Process
• define the partially observable Markov Decision Process and contrast it with a regular Markov Decision Process
• describe how the value iteration algorithm is used with the partially observable Markov Decision Process
• describe how a partially observable Markov Decision Process can be implemented with an intelligent agent
• describe the Markov Decision Process and how it can be used by an intelligent agent