Table of Contents

• Lectures
• Chapter homeworks
• NN-using-TF
• Project
• Schedule
• Example posters

Sections in the main syllabus (this) page:

• Instructor & contact information
• Course description in UMSL catalog
• Prerequisites
• Learning outcomes
• Textbook
• Course topics
• Academic honesty
• Programming language
• Due dates and late policy
• Homeworks
• Concept-map homeworks
• Quizzes
• In-person tests
• Grade composition
• Grading scheme
• Instructions for taking a quiz using Proctrio
• Using Overleaf
• Resources
• What other students (who took this course) say

• Instructor (Facilitator): Badri Adhikari
• Email: [email protected]
• Class meets: 6:55 to 8:10 PM - Mondays at SSB 410 and Wednesdays online synchronously via Zoom (see Canvas for the link)
• Office hours: Mondays 4:55 PM to 6:55 PM (please email to inform)

Teaching philosophy: Computer science and technology is mostly a practical discipline. To learn the fundamentals, an effective strategy is to follow an iterative process of reading, analyzing, and coding. However, many students either like to analyze or code but not both. While some of us enjoy developing the skills for critically assessing the concepts and algorithms, many others enjoy programming and love building things. I think that an effective computer science course should be a balance of (a) theoretical knowledge to understand how computer technology works, and (b) implementation skills to test and execute the theories and algorithms. I design course contents and assignments so that students have an opportunity to improve both: analytical and programming skills. Students with a rich programming experience may find this balance slightly easier but will have a platform to explore further. For many others who do not consider themselves expert programmers, taking such a course can become a rewarding experience.

This course provides an introduction to artificial intelligence (AI). The list of topics may include artificial neural networks, search, planning, knowledge-based reasoning, probabilistic inference, machine learning, natural language processing, and practical applications. [3 credit units].

CMPSCI 3130 (Design and Analysis of Algorithms) or Graduate Standing in CS.

This course consists of (a) the book's chapter topics in the form of recorded lectures (see lectures tab) along with corresponding homeworks, (b) a collection of activities as a crash course on 'machine learning using Keras/Tensorflow' (see nn-tf tab), and (c) a semester-long project on feed-forward neural networks (see project tab). If you are new to artificial intelligence, Python, and machine learning, the activities and the project will require some exploration and study. But they provide you opportunities to learn the fundamentals of neural networks and the state-of-the art libraries to build prediction systems. Below are the learning outcomes:

• Use Python, Numpy and Keras to design, train, and evaluate basic feed-forward neural networks
• Learn artificial intelligence principles and approaches
• Learn a basic understanding of the building blocks of AI as presented in terms of intelligent agents
• Evaluate various searching algorithms
• Understand problems and ideas in the field of natural language processing, perception, and robotics
• Learn the philosophical foundations of AI and the future of AI
• Implement various AI algorithms such as depth-first search and breadth-first search.

Artificial Intelligence: A Modern Approach (3rd Edition) by Pearson.

• Book chapters

  • Chapter 18 - Learning From Examples
  • Chapter 1 - Introduction
  • Chapter 2 - Intelligent Agents
  • Chapter 3 - Solving Problems by Searching
  • Chapter 5 - Adversarial Search
  • Chapter 7 - Logical Agents
  • Chapter 22 - Natural Language Processing
  • Chapter 24 - Perception
  • Chapter 25 - Robotics
  • Chapter 26 - Philosophy, Ethics, and Safety of AI

• Machine learning using Tensorflow

  • Basics of Python, Numpy and Keras
  • Design, train, and evaluate basic feed-forward neural networks
  • Study feature importance and feature reduction

Any form of academic dishonesty in this class will result in an F for the semester and the case will be referred to the provost's office for possible further disciplinary action, regardless of how trivial it is. Please don't use another student's assignment (or a solution in the internet) to complete your own assignment. Discussing the material is 'OK', but please do your work on your own. You should complete the homework alone, not together, and not in a group. If you have any questions about any of the lessons or the assignments, please contact me, and I will point you in the right direction. Please read UMSL's policy and keep yourself out of plagiarism. In your reports, all sources must be clearly cited. You should not copy-paste any content from the internet without citing. "If you didn’t write it (or create it), cite it". If you have not written an original report in the past, please read "Best practices to avoid plagiarism" and "9 things you should already know about plagiarism" before working on your report. Also, please note that our turnitin tool automatically checks for plagiarism.

Python3 is the programming language for this course; you are expected to use Python3 for all of your classroom activities, homeworks, and the project. You are welcome to use Google colab or your own hosted Jupyter Notebook for running your programs.

• Quizzes, homeworks and project phases have their respective due dates (see Schedule and see Canvas).
• You can request a maximum two-day extension on any homeworks or project submissions - for up to two submissions.
• If you email me a few hours before a deadline and I don't reply you immediately, and if you have not used your two-day extensions, you can assume that the extension is granted automatically.
• Once you use your extension days, late submissions will get no points.

There will be three types of homeworks: project homeworks (see project tab), drawing concept maps as chapter summaries (see concept-map tab), and some chapters have additional homeworks (see chapter-homeworks tab). All homeworks should be submitted via Canvas.

The homework here is to make a concept map for each chapter after watching the lectures in the 'lectures' page. Whenever possible, your concept map should be divided into the 'topics of the chapter' (see the lectures page). In order to receive full points, each concept map should contain all of the key concepts/ideas discussed in the lectures. Two commonly used tools to make concept maps are Lucidchart and Slatebox. You are welcome to use any other resources, including a pen & a paper or an electronic pen and writing tools. Concept maps should be submitted to the respective discussion boards so they are visible to other students in the class. You are requires to submit concept maps in .jpg, .jpeg, or .png format. You can view the concept maps uploaded by other students in the class only after you have submitted yours. The following two videos explain how to make your first concept map.

1. How to make the perfect mind map and study effectively?
2. How to make a concept map

Also, here are some example concept maps drawn by other UMSL students:

1. Philosophical foundations by Miguel Corona
2. Fair AI by Fiyanshu Arora
3. P vs NP vs NPC by Jacob Barger

In addition to drawing a concept map, after watching the lectures in a chapter, you will also need to take a five minute quiz. The questions on the quiz will be multiple-choice or true/false type. Please take this quiz soon after watching the chapter lectures. Also, before taking the first quiz, please read the instructions on proctoring (see proctoring tab).

There will be two in-person tests in this course (mid-term and final). Questions in the test will be similar to the homework questions.

Note: You should submit the course evaluation survey at the end of the semester to receive your final grade.

Will be posted soon! Smartproctoring: https://keeplearning.umsystem.edu/students/other-tools/smarterproctoring

All reports should be prepared using Overleaf. You are welcome to use any templates you want. Here is an example. Please learn more about Overleaf here.

Some of you may have wondered why this course requires you to prepare project reports in Overleaf and why Microsoft Word, Open Office Writer, or Google doc are not allowed. Below are some bullets that capture my thoughts on why learning Overleaf/Latex is useful, particularly while you are in college. Please remember that fundamentally, Latex/Overleaf is typesetting system while Word or Writer are word processors.

• From a CS conference and journal publisher's point of view giving authors a LaTeX template ensures that all the papers have a uniform formatting, and that you won't spend much time fixing formatting issues. This saves publication costs.
• Papers written in Latex have the same output whether the authors prepared them on a Linux, a Mac or Windows machine. In OpenOffice Writer or Word, this is not always the case.
• As your document size increases, particularly if they contain many images, softwares such as Word and Writer start to crash. Latex/Overleaf may take slightly longer to process, but they usually don't crash. It's extremely stable, no matter how complex the documents are. At least, you don't have to worry about crashing when typing/preparing your content.
• Similar to OpenOffice Latex is free. Please note that Overleaf may not be free in future but the Latex code can be compiled in free Latex compilers outside of Overleaf.
• Google doc is a great tool for collaborative efforts but it lacks many features that Latex/Word offer. For example, citing/organizing references. Latex has excellent referencing system.
• Some top conferences on Artificial Intelligence and Machine Learning (such as ICML) only accept submissions in Latex format.
• Many CS/IT graduates who have prepared reports using Overleaf/Latex often brag about high typographical quality of their reports. This is particularly true for documents that are heavy on mathematics. Those who don't know about Latex, always keep wondering how Word/Writer can be used to prepare such reports.
• TeX (The core of Overleaf/Latex) has been around for over thirty years, and the underlying language hasn't changed very much in that time.
• Your document is relatively safe because the file format is open and there's lesser virus threat.
• Latex provides a consistency of the layout, i.e., it is really difficult to mess up the typography. This lets you concentrate on the contents and does not distract you by being concerned about the looks of the document.

I am not an Overleaf/Latex maniac and I don't use it for everything. However, I strongly believe that CS/IT students should learn it once while at university. Even just to know that such an alternative to Word/Writer exists and that they are widely used.

Your success in this class is important to me. If you need official accommodations, you have a right to have these met. If there are aspects of this course that prevent you from learning or exclude you, please let me know as soon as possible. Together we’ll develop strategies to meet both your needs and the requirements of the course.

"Sorry to bother you after the course is done. Thank you for your teaching and hard work for this semester. I enjoy your teaching in this course, and I learned a lot through this course. When starting this course, I have no experience with python and TensorFlow, but I can build my neural network right now. This gives me a ton of experience in AI and data science which I believe is very useful for my career. Also, through this course, I can feel the charm of AI. I get attracted to it, and I want to learn more about it when I am going to graduate school." - A student in fall 2021 class (online).

"The hands-on approach of the activities and the course project were the best part of the course. These activities permitted us to delve as deep as we want in understanding the concepts. The course project allows us to use all the ideas learned from the activities and apply them to a problem of our choosing." - A student in fall 2020 class (on-line).

"This course is like no other in the computer science department. Professor Badri went above and beyond to help us achieve the skills we will use in our work life after university. Many professors keep teaching the same theory stuff again and again and after the semester its impossible to recall what we studied but for this course we did a lot of programming and that we will remember and use in our actual life. I loved this class and other classes taught in this way." - A student in spring 2020 class (online).

"I loved this course. Separating theory and programming was an excellent idea. I wish it was done for all courses." - A student in spring 2020 class (in-person).

"I particularly liked all of these resources he provided to help us learn and guide us through the course." - A student in fall 2019 class (in-person).

"I had not done Python programming before so I was bit lost at the beginning but the activity video lectures and sample Python notebooks helped me excel in the course. Being able to see the project report done by other students who had taken the course last semester was nice." - A student in fall 2019 class (in-person).