CSC 301.01, Class 41: Wrapup

Overview

• Preliminaries
  • Notes and news
  • Upcoming work
  • Extra credit
  • Questions
• More about the final
• Some learning goals/outcomes

News / Etc.

• Please attend Friday’s class.
• I’ve posted a sample final.

Upcoming work

• Homework 10 due TONIGHT.

Extra Credit (Academic/Artistic)

Extra credit (Peer)

• Pub-free quiz, TONIGHT.

Extra Credit (Misc)

Other good things

• Musical this weekend.
• More music stuff.

Questions

Should I submit really old extra credit writeups that are sitting on my computer?

It couldn’t hurt.

Will you be available next week?

I should be sitting in my office much of the day on Monday and Tuesday.

More about the final

• Goal: A chance for you to demonstrate what you’ve learned/mastered. (See below.)
• Often, more “can you read and apply/critique an algorithm” than “can you design one”?
• We’ll look quickly at the sample final.

Some learning goals/outcomes

Note: In spite of what I said about EOCEs, we’re really doing this because I think it’s useful to step back and look at what you learned (or might have learned).

Big picture:

• Skills and strategies for designing algorithms and data structures.
• Increased knowledge of “the literature” of common algorithms and data structures.
• Other related stuff

Habits of thought in dealing with algorithm design

• Consider approaches
• Develop a potential algorithms to solve a problem
• Work though examples
• Try to find a counter-example
• Try to prove it correct
• Analyze it
• Do it all over again (“Can I do better?”)

Approaches to algorithm design

• Use one of the main techniques
  • Greed
  • Divide and conquer
  • Exhaustive search (and dynamic programming)
  • Turn the problem into an already solved problem
• Plus …
  • Drawing pictures can help
  • Modify an existing algorithm
  • Introduce randomness
  • Restrict the input
  • Accept approximations

“The literature” - Classic (and not so classic) algorithms and data structures

Algorithms

• Rabin-Karp - Substring matching based on hashing
• Shortest path - Solve the more general problem: Find the shotest path from s to every node. (Dijkstra’s)
• Prim - Minimum spanning tree (builds connected)
• Kruskal - Minimum spanning tree (builds disconnected)
• Ford-Fulkerson - Max flow
• Maximum bipartite
• Heap sort and other sorting algorithms
• Min stamps
• Approximate matching
• 0-1 knapsack
• Plus many many more.
• Heapsort

ADTs

• Lists
• Dictionary/Map
• Set
• Graph
• Tree

Data Structures

• Skip list
• Trie (there are multiple implementations of tries, but we usually use tries as a way to achieve an ADT or other goal)
  • Used for those weird suffix tree things useful for pattern matching
  • Dictionaries
• Balanced search trees
  • Red-black trees
  • 2:3 trees
  • B tree (generalized 2:3 trees)
• Heaps
• Hash tables
• Union-find
• Different implementations of graphs
  • Adjacency list
  • Adjacency matrix

Other stuff

• Formalized asymptotic analysis - Big O, Big Omega, Big Theta
  • A bit of amortized analysis
• Techniques for solving recurrence relations.
  • Master theorem
  • Recurrence trees
  • Repeated expansion
• Way too many summations and rules for summations.
• Program verification
  • Memory/state models
  • Loop invariants
• Proof, informal and formal
  • Induction (strong and weak)
  • Constructive
  • Contradiction
  • …
• Counter-examples (covered earlier, but …)
• Practice with (stupid) hackerrank approaches
• Practice with C and GNU code conventions