Scribd
Upload
39 views

Dynamic Programming: The Basics

This document discusses dynamic programming and its application to solving shortest path problems. It provides an example of using dynamic programming to find the shortest distance between nodes in a network by breaking the problem down into stages. It explains the computational advantages of the dynamic programming approach over total enumeration. Finally, it outlines some key characteristics of dynamic programming problems and lists some other examples that will be solved.

Uploaded by

Venkatesh Gopalaswamy
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
39 views

Dynamic Programming: The Basics

This document discusses dynamic programming and its application to solving shortest path problems. It provides an example of using dynamic programming to find the shortest distance between nodes in a network by breaking the problem down into stages. It explains the computational advantages of the dynamic programming approach over total enumeration. Finally, it outlines some key characteristics of dynamic programming problems and lists some other examples that will be solved.

Uploaded by

Venkatesh Gopalaswamy
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 12

DYNAMIC PROGRAMMING

The Basics
Shortest Path Problem
 1st Problem
SHORTEST PATH PROBLEM
Stage 5
 What is the shortest distance from node
10 to itself?
 f5(10) = 0!
Stage 4
 What is the shortest distance from node
8 to node 10?
 f4(8) = 300 + 0 = 300. r4(8) = (8,10).
 What is the shortest distance from node
9 to node 10?
 f4(9) = 350 + 0 = 350. r4(9) = (9,10).
Stage 3
 What is the shortest distance from node 5 to node
10?
 f3(5) = Min {780+300, 600+350} = 950. r3(5) = (5,9).
 What is the shortest distance from node 6 to node
10?
 f3(6) = Min {240+300, 360+350} = 540. r3(6) = (6,8).
 What is the shortest distance from node 7 to node
10?
 f3(7) = Min {250+300, 600+350} = 550. r3(7) = (7,8).
Stage 2
 What is the shortest distance from node 2 to node
10?
 f2(2) = 1360. r2(2) = (2,6).
 What is the shortest distance from node 3 to node
10?
 f2(3) = 990. r2(3) = (3,6).
 What is the shortest distance from node 4 to node
10?
 f2(4) = 950. r2(4) = (4,7).
Stage 1
 What is the shortest distance from node
1 to node 10?
 f1(1) = Min {510+1360, 600+990,
1000+950} = 1590.
 r1(1) = (1,3).
Computational Advantage of
DP
 If you were to enumerate all paths, how many are
there?
 18 (How?)
 What is the total computational effort with total
enumeration?
 54 additions and 17 comparisons (How?)
 What is the total computational effort with DP?
 18 additions, 11 comparisons (How?)
 Supposing that the shortest path problem had 11
stages, with 10 states each in stages 2 to 10.
 How do the 2 approaches now compare?
Characteristics of DP
 Problem can be divided into stages with a
decision required at each stage.
 Each stage has a number of states
associated with it.
 A decision chosen at the current stage
transforms a state in the current stage to
some other state in the next stage.
Characteristics of DP
 Given the current state in stage t, the optimal
decision in each of the remaining stages does
not depend on previously reached states or
previously chosen decisions.
 PRINCIPLE OF OPTIMALITY
 A recursion relates the cost (reward) in stage t
to the cost (reward) earned from stages t+1 to
T.
ft(i) = Min j{cij + ft+1(j)}
Other DP Problems
 We will now solve each of the remaining
problems
 INVENTORY PLANNING AT XYZ INC. -
Deterministic Case
 INVENTORY PLANNING AT XYZ INC. - Stochastic
Case
 OPTIMAL REPLACEMENT POLICY AT BHARAT
TOOLS – finite time horizon
 OPTIMAL TIMING OF LIQUIDATION OF STOCK

You might also like