Chapter 18: Artificial Intelligence

Posted on Edited on In , Views: Disqus:

Definitions:

Supervised Learning:

  • Using a large number of tasks with given outcomes to enable a computer program to improve its performance in accomplishing similar tasks

Unsupervised learning:

  • Using a large number of tasks with unknown outcomes to enable a computer program to improve its performance in accomplishing similar tasks

Reinforcement learning:

  • Using a large number of tasks with unknwon outcomes and use of feedback to enable a computer program to improve its performance in accomplishing similar tasks

Graph:

  • A graph is a collection of nodes or vertices between which there can be edges.
  • Each node has a name
  • An edge can have an associated label which is a numerical value

The use of graph to aid Artificial Intelligence:

  • Artificial neural networks can be represented by graphs
  • Graphs provide structure for relationships // graphs provide relationships between nodes
  • AI problem can be defined as finding a path in a graph
  • Graph may be analysed/injected by a range of algorithms
    • A* / Dijkstra’s algorithm
    • used in machine learning
  • Back propagation of errors / regressions

State the reasons for having multiple hidden layers in an artificial neural network

  • Enable deep learning to take place
  • Where the problem you are trying to solve has a higher level of complexity it requires more layers
  • Allow neural network to learn and make decisions on its own
  • To improve the accuracy of the result

Explain how artificial neural network enable machine learning

  • Artifical neural networks are intended to replicate the way human brain work
  • Weights / values are assigned for each connection between nodes
  • The data are input at the input layer, then passed into the system
  • They are analyzed at each subsequent hidden layer where characteristics are extracted / output are calculated
  • This process of training / learning is repeated mant times to achieve optimum outputs // reinforcement learning takes place
  • Decisions can be made without being specifically programmed
  • The output layer provides the result
  • Back propagation of errors will be used to correct any errors that have been made