CS/SE310 Analysis of Algorithms

Fall 2019

Instructor Noman Nazar

Lectures Days: Wed, Sat Contact Email: [email protected]

Time: 3:30 – 4:45

Office Hall STD404, 2nd Floor, TA <Name> TBA

STD Buiding, UMT <Office Hours and Venue> TBA

Office Hours: TBA

Lab NA

Pre-  CS/SE200-Discrete Moodle

Requisites Mathematics Group
 CS/SE210-Data Structures
and Algorithms

Course Course objectives are the following, in order of importance:-

Objectives  To familiarize the students with techniques of analysis
 To enable the students to choose and apply a relevant analysis technique for
an algorithm they encounter
 To impart design ideas that go hand-in-hand with the analysis of algorithms
 To familiarize the students with major paradigms of algorithms
 To familiarize the students with commonly employed algorithms and their
analysis

Textbooks  Cormen, Leisorson, Rivest & Stein, “Introduction To Algorithms”, 2nd

Edition, Prentice Hall of India

Technology  C/C++/Java

Midterms Single Midterm Examination Final Single Final Examination that will
from the material covered be comprehensive, i.e., will be
during the first 7 weeks assessing the whole course
contents

Classroom  Review recommended material before coming to class

Policy  No disturbances
 Keep mobiles switched off.
 Plagiarism will result in negative marking

Grading  Assignments: 5 in number, 20%

Policy  Quizzes: 6 in number, 20%
 Midterms: 1 in number, 20%
 Final: 1 in number, 40%
Tentative Course Plan:
Week Contents
1  What is an algorithm?
 Few examples of algorithms
 Role of algorithms in computing
 Implementation of algorithms and technology
 Need for efficiency
 Example of how efficiency makes a difference
 Expressing an algorithm
 Pseudocode conventions
 Few examples on pseudocoded algorithms
o Evenness/Oddness of a number
o Finding if a number divides another number
o Finding LCM of two numbers
2  What is analysis of an algorithm?
 Factors affecting analysis
 Counting steps and examples
o Generating Primes
o Checking whether a number is a Perfect Number
 Estimate from costly operations
 Variations inherent in algorithms
o Example: Searching a List/Array
 Need for Worst/Average/Best cases
 Need for envelopes/families of time complexity
 The Order concept
 Big-O notation and definition
 Big-O examples applied on algebraic expressions
o Quadratic
o Cubic
o Polynomial
o Logarithmic
o Rule of thumb
3  Graphing the time values for an algorithm
 Relation between discrete expressions and their continuous counterparts
 Trends extractable from the continuous counterparts
 Asymptotic analysis
 Big-O revised
 Theta and Omega notations and definitions
 Few examples applied on algebraic expressions
 Asymptotic trends illustrated with graphs
 Sorting algorithms, their common use and rich variety of algorithmic principles
o Selection Sort and analysis
 Bubble Sort and analysis
 Insertion Sort and analysis using Tabular approach
4  Shifting the approach: Divide and Conquer
 Principle of Merging
 Merge algorithm and complexity
 Terminology introduced: in-place/out-place; auxiliary memory
 Merge Sort explanation with diagrams
 Estimate of complexity
 Merge Sort algorithm
 Estimate of complexity of Merge Sort using Tabular approach
 Reviewing Divide and Conquer features encountered in Merge Sort
  General features of Divide and Conquer problems
 Limitations of Divide and Conquer approach
 Revisiting Merge Sort complexity and forming a recurrence based on Divide
and Conquer features
 Solving the recurrence using iteration method
5  Recurrences and their relation to Divide and Conquer algorithms
 Solving recurrences: Iteration method
o 2-3 examples
 Solving recurrences: Recursion trees
o 2-3 examples
 Solving recurrences: Master’s theorem
o 4-5 examples
 Preparing for Quick Sort: Concept of a Pivot and Partition
6  Partition algorithm
 Quick Sort algorithm
 Worst and Average cases of Quick Sort
 Partition analysis
 Quick Sort analysis
 Variations on Quick Sort
7  Shifting the approach: Linear Time Sorting
 The Counting Principle
 Counting Sort and analysis
 The Bucket Principle
 Bucket Sort and analysis
 The use of Radix as basis for bucket
 Radix Sort and analysis
 Revision
8  Mid Term Examination
 Shifting the approach: Greedy algorithms
 Example: Fractional Knapsack problem and solution
 Features of greedy algorithms evident in knapsack problem
 Concepts of solution space and optimal solution
 General structure of greedy algorithms
 Example: Shortest job first problem
9  Greedy algorithm example: Huffman codes
 Where greedy algorithms fail through example of making change
 Analysis of failure
 Shifting the approach: Dynamic Programming
 Dynamic programming introduced through making change problem’s alternate
solution
 Examples on dynamic programming: knapsack revisited
 Matrix Chain Multiplication
o Multiplying matrices in a chain
o Costs incurred
o Similarity to putting Parentheses
o Data Structures to be used
o The DP algorithm approach
10  Matrix chain algorithm pseudocode
 Filling the DP table
 Tracing the optimal solution
 Dynamic Programming applicability in general and features of DP problems
 Shifting the approach: Amortized Analysis
  Amortized analysis introduced through example of a binary counter
 Aggregate analysis
11  Amortized analysis: Other methods
o Accounting method
o Potential method
 Other examples
o Multistacks
o Dynamic tables
 Graphs
 Representing graphs
o Matrix
o List
 Common types of graphs
o Simple/Multi/Pseudo
o Directed/Undirected
o Weighted/Unweighted
 Example of elementary graph algorithm: Connectedness
 Graph Searches and Traversals
12  Breadth-First Search
 Properties
 Depth-first Search
 Recursive and Non-recursive representations
 Properties
 Classification of Edges
13  DFS applications
o Topological Sorting
o Strongly-connected Components
 Minimal Spanning Trees
 Kruskal’s algorithm
 Prim’s algorithm
14  Shortest Paths
 Dijkstra’s algorithm
 Floyd-Warshall algorithm
 Revision
15  Shifting the approach: Randomized algorithms
 Example: Hiring problem and analysis
 Random walks
 Monte-Carlo algorithms
 Example: Fission modeling, Probable primes
 Las Vegas algorithms
 Example: Verifying Matrix products