Lecture 1 - Multi Liber

Lecture 1 - Multi Liber

Lecture 3 Note: Some slides and/or pictures are adapted from Lecture slides / Books of Dr Zafar Alvi. Text Book - Aritificial Intelligence Illuminated by Ben Coppin, Narosa Publishers. Ref Books

Artificial Intelligence- Structures & Strategies for Complex Problem Solving by George F. Luger, 4th edition, Pearson Education. Artificial Intelligence A Modern Approach by Stuart Russell & Peter Norvig. Artificial Intelligence, Third Edition by Patrick Henry Winston Outline

Problems and their representation Goal Driven VS Data Driven

Properties of search Methods Tree search algorithm Depth First algorithm Breadth First algorithm Iterative Deepening algorithm Problems and their Representations

Three men and three lions are on one side of a river, with a boat. They all want to get to the other side of the river. The boat can only hold one or two at a time. At no time should there be more lions than men on either side of the river, as this would probably result in the men being eaten.

Representation could be 3, 3, 1 0, 0, 0 Problems and their Representations Traveling Salesman problem (NP complete) Problems and their Representations Traversing a Maze

Data Driven or Goal Driven Search Data-driven search starts from an initial state and uses actions that are allowed to move forward until a goal is reached. This approach is also known as forward chaining. Search can start at the goal and work back

toward a start state, by seeing what moves could have led to the goal state. This is goaldriven search, also known as backward chaining. Properties of Search Methods

Complexity Completeness Optimality

Admissibility Irrevocability Complexity It is useful to describe how efficient that method is, over time and space. The time complexity of a method is related to

the length of time that the method would take to find a goal state. The space complexity is related to the amount of memory that the method needs to use. Completeness A search method is described as being

complete if it is guaranteed to find a goal state if one exists. A method that is not complete has the disadvantage that it cannot necessarily be believed if it reports that no solution exists. Optimality

A search method is optimal if it is guaranteed to find the best solution that exists. In other words, it will find the path to a goal state that involves taking the least number of steps. This does not mean that the search method itself is efficientit might take a great deal of time for an optimal search method to identify the optimal

solutionbut once it has found the solution, it is guaranteed to be the best one. Admissibility In some cases, the word optimal is used to describe an algorithm that finds a solution in the quickest possible time, in which case the

concept of admissibility is used in place of optimality. An algorithm is then defined as admissible if it is guaranteed to find the best solution. Irrevocability Methods that do not use backtracking, and

which therefore examine just one path, are described as irrevocable. Irrevocable search methods will often find suboptimal solutions to problems because they tend to be fooled by local optima solutions that look good locally but are less favorable when compared with other solutions

elsewhere in the search space. Tree search algorithms Basic idea: offline, simulated exploration of state space by generating successors of already-explored

Tree search example Tree search example Tree search example Uninformed search strategies

Uninformed search strategies use only the information available in the problem definition Depth-first search Breadth-first search Iterative deepening search etc

Depth-first search Expand deepest unexpanded node Implementation: put successors at front Depth-first search

Expand deepest unexpanded node Implementation: put successors at front Depth-first search Expand deepest unexpanded node

Implementation: put successors at front Depth-first search Expand deepest unexpanded node Implementation:

put successors at front Depth-first search Expand deepest unexpanded node Implementation: put successors at front

Depth-first search Expand deepest unexpanded node Implementation: fringe = LIFO queue, i.e., put successors at front

Depth-first search Expand deepest unexpanded node Implementation: fringe = LIFO queue, i.e., put successors at front

Depth-first search Expand deepest unexpanded node Implementation: put successors at front Depth-first search

Expand deepest unexpanded node Implementation: put successors at front Depth-first search Expand deepest unexpanded node

Implementation: put successors at front Depth-first search Expand deepest unexpanded node Implementation:

put successors at front Depth-first search Expand deepest unexpanded node Implementation: put successors at front

Depth-first search Function depth () { queue = []; // initialize an empty queue state = root_node; // initialize the start state

while (true) { if is_goal (state) then return SUCCESS else add_to_front_of_queue (successors (state)); if queue == []

then report FAILURE; state = queue [0]; // state = first item in queue remove_first_item_from (queue); } }

Breadth-first search Expand shallowest unexpanded node Implementation: new successors go at end Breadth-first search

Expand shallowest unexpanded node Implementation: new successors go at end Breadth-first search Expand shallowest unexpanded node

Implementation: new successors go at end Breadth-first search Expand shallowest unexpanded node Implementation:

fringe is a FIFO queue, i.e., new successors go at end Breadth-first search Function breadth () { queue = []; // initialize an empty queue

state = root_node; // initialize the start state while (true) { if is_goal (state) then return SUCCESS else add_to_back_of_queue (successors (state)); if queue == []

then report FAILURE; state = queue [0]; // state = first item in queue remove_first_item_from (queue); } } Iterative deepening search l =0

Iterative deepening search l =1 Iterative deepening search l =2 Iterative deepening search l =3

Iterative deepening search Number of nodes generated in a depth-limited search to depth d with branching factor b: NDLS = b0 + b1 + b2 + + bd-2 + bd-1 + bd GP Number of nodes generated in an iterative deepening search to depth d

with branching factor b: NIDS = (d+1)b0 + d b^1 + (d-1)b^2 + + 3bd-2 +2bd-1 + 1bd For b = 10, d = 5, NDLS = 1 + 10 + 100 + 1,000 + 10,000 + 100,000 = 111,111 NIDS = 6 + 50 + 400 + 3,000 + 20,000 + 100,000 = 123,456

Overhead = (123,456 - 111,111)/111,111 = 11%

Recently Viewed Presentations

  • Olmec - Political

    Olmec - Political

    Aztec - Society. Social hierarchy. Emperor Nobles: military leaders, gov't officials, and priests. Commoners. Slaves. Aztec - Political. Empire of tribute states. Aztecs took over neighboring areas…allowed local rulers to stay in power as long as they paid tribute.
  • The Nation State

    The Nation State

    Centripetal forces include external aggression, which may stimulate national bonding against a common enemy; federal structures that allow the safe expression of regional and other subnational differences; and national mass media and education, which create a shared culture and language....
  • Primer on Addiction Care in the Family Medicine Setting

    Primer on Addiction Care in the Family Medicine Setting

    Furosemide 43.4 . 14. Metoprolol 38.9 . 15. Zolpidem Drug of abuse 38 21. Oxycodone/Acetaminophen Drug of abuse 31.9 A Billion Dollar Industry (2012 Sales) Oxycontin $2.695 billion. Suboxone $1.349 billion ... It has greater safety in overdose.
  • Harvard Outline - Mesa Public Schools

    Harvard Outline - Mesa Public Schools

    Why create an outline? To show the hierarchical relationship or logical ordering of information. For research papers outlines help keep track of large amounts of information. For creative writing outlines organize various plot threads and character traits. For an oral...
  • ODE WBSCM RA ordering/reports manual

    ODE WBSCM RA ordering/reports manual

    Running A Value of Commodities Received - RA Report The Value of Commodities Received Report shows schools the Value of the USDA Foods they have received during a certain time period. Business Managers often ask for this information at the...
  • Undergoers and goals: on measuring out and being affected

    Undergoers and goals: on measuring out and being affected

    Undergoers and goals: on measuring out and being affected. Workshop on Affectedness, NTU, June 17-20th. Boban Arsenijević, University of Niš
  • UTA Presentation - Smarter Transportation

    UTA Presentation - Smarter Transportation

    Confidential | 2016. Industries We Serve. Services We Provide. We improve moments in millions of lives in 18 Industry Sectors… and deliver value to our clients by providing the broadest portfolio of Services.
  • Genigraphics Research Poster Template 36x56

    Genigraphics Research Poster Template 36x56

    The adults were presented word and sentence lists to read. The children were shown picture plates and asked to say or repeat the name of the picture. Children were read the storybook (Brown bear, brown bear) and asked to repeat...