About the Book
I find that I don’t understand things unless I try to program them.
—Donald E. Knuth, The Art of Computer Programming, Volume 4
There are many excellent books on Algorithms — why in the world we would write another one??? Because we feel that while these books excel in introducing algorithmic ideas, they have not yet succeeded in teaching you how to implement algorithms, the crucial computer science skill.
Our goal is to develop an Intelligent Tutoring System for learning algorithms through programming that can compete with the best professors in a traditional classroom. This MOOC book is the first step towards this goal written specifically for our Massive Open Online Courses (MOOCs) forming a specialization “Algorithms and Data Structures” on Coursera platform and a microMasters program on edX platform. Since the launch of our MOOCs in 2016, hundreds of thousands students enrolled in this specialization and tried to solve more than hundred algorithmic programming challenges to pass it. And some of them even got offers from small companies like Google after completing our specialization!
In the last few years, some professors expressed concerns about the pedagogical quality of MOOCs and even called them the “junk food of ed- ucation.” In contrast, we are among the growing group of professors who believe that traditional classes, that pack hundreds of students in a single classroom, represent junk food of education. In a large classroom, once a student takes a wrong turn, there are limited opportunities to ask a question, resulting in a learning breakdown, or the inability to progress further without individual guidance. Furthermore, the majority of time a student invests in an Algorithms course is spent completing assignments outside the classroom. That is why we stopped giving lectures in our offline classes (and we haven’t got fired yet :-). Instead, we give flipped classes where students watch our recorded lectures, solve algorithmic puzzles, complete programming challenges using our automated homework checking system before the class, and come to class prepared to discuss their learning breakdowns with us.
When a student suffers a learning breakdown, that student needs immediate help in order to proceed. Traditional textbooks do not provide such help, but our automated grading system described in this MOOC book does! Algorithms is a unique discipline in that students’ ability to program provides the opportunity to automatically check their knowledge through coding challenges. These coding challenges are far superior to traditional quizzes that barely check whether a student fell asleep. Indeed, to implement a complex algorithm, the student must possess a deep understanding of its underlying algorithmic ideas.
We believe that a large portion of grading in thousands of Algorithms courses taught at various universities each year can be consolidated into a single automated system available at all universities. It did not escape our attention that many professors teaching algorithms have implemented their own custom-made systems for grading student programs, an illustration of academic inefficiency and lack of cooperation between various instructors. Our goal is to build a repository of algorithmic programming challenges, thus allowing professors to focus on teaching. We have already invested thousands of hours into building such a system and thousands students in our MOOCs tested it. Below we briefly describe how it works.
When you face a programming challenge, your goal is to implement a fast and memory-efficient algorithm for its solution. Solving programming challenges will help you better understand various algorithms and may even land you a job since many high-tech companies ask applicants to solve programming challenges during the interviews. Your implementation will be checked automatically against many carefully selected tests to verify that it always produces a correct answer and fits into the time and memory constrains. Our system will teach you to write programs that work correctly on all of our test datasets rather than on some of them. This is an important skill since failing to thoroughly test your programs leads to undetected bugs that frustrate your boss, your colleagues, and, most importantly, users of your programs.
You maybe wondering why it took thousands of hours to develop such a system. First, we had to build a Compendium of Learning Breakdowns for each programming challenge, 10–15 most frequent errors that students make while solving it. Afterwards, we had to develop test cases for each learning breakdown in each programming challenge, over 20 000 test cases for just 100 programming challenges in our specialization.
We encourage you to sign up for our Algorithms and Data Structures specialization on Coursera or MicroMasters program on edX and start interacting with thousands of talented students from around the world who are learning algorithms. Thank you for joining us!
About the Authors
Alexander S. Kulikov is a senior research fellow at Steklov Mathematical Institute of the Russian Academy of Sciences, Saint Petersburg, Russia and a lecturer at the Department of Computer Science and Engineering at University of California, San Diego, USA. He also directs the Computer Science Center in Saint Petersburg that provides free advanced computer science courses complementing the standard university curricula. Alexander holds a Ph. D. from Steklov Mathematical Institute. His research interests include algorithms and complexity theory. He co-authored online courses "Data Structures and Algorithms" and "Introduction to Discrete Mathematics for Computer Science" that are available at Coursera and edX.
Pavel Pevzner (http://cseweb.ucsd.edu/~ppevzner/) is Professor of Computer Science and Engineering at University of California San Diego (UCSD), where he holds the Ronald R. Taylor Chair and has taught a Bioinformatics Algorithms course for the last 12 years. In 2006, he was named a Howard Hughes Medical Institute Professor. In 2011, he founded the Algorithmic Biology Laboratory in St. Petersburg, Russia, which develops online bioinformatics platform Rosalind (http://rosalind.info). His research concerns the creation of bioinformatics algorithms for analyzing genome rearrangements, DNA sequencing, and computational proteomics. He authored Computational Molecular Biology (The MIT Press, 2000), co-authored (jointly with Neil Jones) An Introduction to Bioinformatics Algorithms (The MIT Press, 2004), and Bioinformatics Algorithms: An Active Learning Approach (Active Learning Publishers, 2014). For his research, he has been named a Fellow of both the Association for Computing Machinery (ACM) and the International Society for Computational Biology (ISCB).