Syllabus
Course Overview¶
Course Description¶
This course will introduce students to the probabilistic and statistical models at the heart of modern artificial intelligence. Specific topics to be covered include probabilistic methods for reasoning and decision-making under uncertainty; inference and learning in Bayesian networks; prediction and planning in Markov Decision Processes; applications to intelligent systems, speech and natural language processing, information retrieval, and robotics.
Prerequisites¶
The course is aimed broadly at advanced undergraduates and beginning graduate students in mathematics, science, and engineering. Prerequisites are elementary probability, linear algebra, and calculus, as well as basic programming ability in Python. Programming assignments should be completed in Python.
Learning Outcomes¶
Upon successful completion of this course, students will be able to:
Describe the structure and behavior of different probabilistic models including Bayes Nets, Hidden Markov Models, Markov Decision Processes, and other similar models.
Perform inference on those models, to derive various quantities from the model parameters.
Prove mathematical relationships between probabilities arising from these models.
Perform learning on those models, to estimate the various parameters from data.
Apply probabilistic models to solve real-world problems.
Design specific models for specific AI tasks.
Implement core algorithms of different models.
Describe how agents learn from data using maximum likelihood learning and reinforcement learning.
How to Succeed in This Class¶
Keep up with lectures and assignments: The pace of this course can pick up quickly, and new concepts often build on earlier material. Staying on top of lectures and assignments each week will help your understanding.
Be consistent: Regularly reviewing course content and completing assignments on time will help you stay on track and prevent any last-minute stress. Make a habit of setting aside dedicated time for this class every week.
Collaborate: I encourage you to talk to your peers, form study groups, and discuss course material to deepen your understanding. Feel free to use Search for Teammates on Piazza to find a study group. Please be sure to follow the collaboration policy outlined in the syllabus.
Attend office hours: Office hours are a great way to get help with course content, ask questions, and connect with the course staff. We have amazing TAs who are eager to help!
Focus on learning over grades: Grades may feel important now, but lasting value is in learning how to think through these problems yourself. Outsourcing the work to a friend, to the internet, or to an AI might get you a grade, but it cheats you out of the skills this course is designed to teach.
CSE 150A / 250A is a rigorous but rewarding course. You will be challenged, but we provide extensive support through office hours and Piazza. Use these resources and you will be well positioned to succeed.
Course Information¶
Instructor: Trevor Bonjour (he/him), tbonjour@ucsd.edu
Please see Course Staff to learn more about the instructional team.
Lectures¶
Lectures will be held in-person at the regularly-scheduled time and place. Attendance is appreciated, but NOT required. Slides/notes will be posted, and classes will be podcasted and posted online for remote viewing.
Lectures for both 150A/250A will be delivered together on
Tue/Thu, 2PM - 3:20PM
, at JEANN
Please note, we have requested that podcasting be enabled for the lectures, but we cannot guarantee that it will always work. You will be able to find the lecture recordings at podcast.ucsd.edu.
Discussions¶
There are no fixed discussion sessions for the course. Each week the homework is released, the TA in charge of the HW will hold a discussion session by the Thursday of that week. You will be notified when the discussion session will happen and the office hours calendar will be updated.
Office Hours¶
Office hours are a drop-in (in person or remote) Question & Answer opportunity to talk with the tutors, TAs, and instructor about concepts and specific questions from class and homework. We offer multiple sessions each week. Please check the office-hours page for the schedule.
Textbook¶
You will not need to purchase any materials for this course. We’ll use lecture slides as our main resource.
The course does not closely follow a particular text. Nevertheless, the following texts (though not required) may be useful as general references. Most of these are freely avaialble (except Russell & Norvig).
Artificial Intelligence: Foundations of Computational Agents (Poole & Mackworth, 2023)
Artificial Intelligence: A Modern Approach (Russell & Norvig, 2020)
Probabilistic Machine Learning (Murphy, 2021)
Reinforcement Learning: An Introduction (Sutton & Barto, 2018)
Mastering Reinforcement Learning (Tim Miller, 2023)
Mathematics for Machine Learning (Deisenroth, Faisal & Ong, 2020)
Getting Started¶
We use several online platforms in this course. Please make sure you have access to all of the following.
Piazza¶
We’ll be using Piazza as our course message board. You will be added to Piazza automatically (Canvas roster sync) if you are registered/waitlisted for the course.
Email may be missed, so Piazza is the best way to ensure timely responses – it is catered to getting you help fast and efficiently from classmates, the TAs, and the instructor. If you need to discuss something privately, you can post a private message to the instructors on Piazza.
If you have a question about anything to do with the course — if you’re stuck on a problem, didn’t understand something from lecture, want clarification on course logistics, or just have a general question about something — you can make a post on Piazza. We only ask that if your question includes some or all of an answer (even if you’re not sure it’s right), please make your post private so that others cannot see it. You can also post anonymously to other students if you prefer.
Course staff will regularly check Piazza and try to answer any questions that you have. You’re also encouraged to answer questions asked by other students. Explaining something is a great way to solidify your understanding of it!
Gradescope¶
We’ll be using Gradescope for homework/project submission and grading. Most of the assignments will be a mixture of math and coding, and the coding parts are usually autograded via Gradescope. You will be automatically (via Canvas roster sync) added to Gradescope if you are registered/waitlisted for the course.
PrairieLearn and PrairieTest¶
We will be using PrairieLearn and PrairieTest in this course for in person assessments. Please see Tests for more information about how to schedule your tests.
Webclicker¶
For in class questions, we’re going to use webclicker.web.app. Please login using your UCSD Google Email address (‘Sign in with Google’) and register for this class using course registration code. If you don’t have a UCSD Google Email (Global Exchange and UCEAP Students), you may use a personal Google account. Enter you PID (usually starting with A) in the Student Identifier field.
The course registration code is KSALDG
.
Canvas¶
We will use Canvas only for posting final grades and for required surveys (such as the #finaid survey and occasional course surveys). All course materials and homework assignments will be available on the course website. Announcements will be made on Piazza. Please check the website for materials and Piazza regularly for announcements and updates.
Course Schedule¶
The course schedule is posted in Schedule.
Typical Week¶
We’ll meet in class on Tuesdays and Thursdays. Homework will be released on Tuesdays and be due on the following Monday. A TA will lead a discussion session every week a homework is released by the Thursday of that week to help you get started on the homework. In addition to the homework, you’ll demonstrate what you learned in two in-person, self-scheduled tests, and you’ll complete an individual mini-project, and an open-ended group project. We recommend talking about the course concepts with other students, and with the instructional team in class, discussions and drop-in office hours.
Assessments¶
Homework Assignments¶
There will be 6 homework assignments. These assignments include written work (proofs and mathematical exercises) and short programs.
Homework assignments will be released on Tuesdays and will be due at 11:59 pm the following Monday.
Submission Format: All homework must be submitted as typed PDFs via Gradescope
. You may use any word processing software (Microsoft Word, Google Docs, LaTeX, etc.). Diagrams may be hand-drawn and scanned, then included in your typed document.
LaTeX Resources: We encourage learning LaTeX as it’s widely used in AI/ML research. LaTeX templates for each assignment will be provided.
For beginners, we recommend Overleaf - a cloud-based LaTeX editor that requires no installation.
This open source LaTeX reference can be helpful when getting started, and you can use the .tex source of all the files we use in class as templates.
Detexify - draw a symbol and it will tell you the LaTeX command for it!
Late Day Policy
There will be no penalty for turning in any of the homework assignments up to 24hrs late. However, we cannot guarantee that late assignments will be graded in a timely fashion. Beyond the 24hr grace period, we will not accept late homework.
Collaboration & AI
We strongly encourage collaboration (but not copying) on homework assignments and projects. Discuss problems together and compare approaches, but each student (or project group) must submit work that reflects their own understanding. Talking through ideas is a great way to learn, but writing your own solutions leads to deeper understanding than simply sharing files.
You may also use AI tools (ChatGPT, Copilot, etc.) in the same spirit: ask questions, get hints, or debug, but don’t copy AI output verbatim. Keep in mind that ChatGPT is infamous for being very confidently wrong, so be critical of its output. Adapt anything you use and make sure you can explain it yourself. Also keep in mind that you won’t have ChatGPT on the tests, so you’ll need to understand the fundamental concepts for yourself in order to do well.
If you have any questions or worries about whether your collaboration constitutes a violation of academic integrity, feel free to ask us on Piazza.
Regrade Requests
Regrades for homework assignments must be requested through Gradescope within three days of the homework assignment being graded and returned to you. Please do not request regrade for homework on Piazza or through email.
If you submit a regrade request, please include a brief but detailed explanation of why you think there was an error in the grading. If you believe your answer is correct and there was an error in grading, please explain that clearly. A valid regrade request must demonstrate that you have reviewed the feedback carefully and provide a clear explanation for the regrade. Requests that do not follow these guidelines will not be considered.
A regrade request may lead to us reviewing the entire assignment and may lead to the score being adjusted up or down depending on any errors found in the original grading. All regrades will be considered once the regrade window closes; thank you in advance for your patience while we carefully look through them.
Projects¶
There are two projects in this course:
Mini-Project (individual): An extended homework-style assignment that walks you through an end-to-end pipeline. The goal is to practice applying concepts on your own before moving to a larger team effort. Submission is individual on Gradescope.
Final Project (group): A deeper, open-ended assignment that serves as the take-home final for the class. You will form groups, select a dataset and problem of interest, and apply methods from the course. The project is due on Gradescope during the official final exam window. Full instructions and rubric will be released later in the quarter.
Please note, there is no grace period for project submissions. We will not accept any late submissions for the projects.
Participation¶
This quarter, for getting instant feedback and measuring participation we will use in-class questions (using Webclicker) and surveys. These are graded for completion not correctness.
In-class questions: You can miss up to two classes with no impact on your participation grade. Any participation points lost beyond that will be redistributed to Test 2.
Surveys: Complete required course surveys when assigned.
Think of this as a chance to engage in class and review your learning.
Tests¶
There will be two in-person, self-scheduled tests. Test 1 will be in Week 4 and Test 2 will be in Week 8. Test 1 covers all material through Week 3 and Test 2 covers all material through Week 7.
Scheduling the Tests
Tests for this course will be administered by the Triton Testing Center (TTC) in the Computer-Based Testing Facility in AP&M B349 and B432
. The TTC’s rules concerning testing are the rules for this course.
You must schedule your tests in advance, and it is recommended that you do so as soon as possible. Scheduling for all tests opens on the first day of instruction.
To make sure you have access to the test questions, you need to be enrolled in PrairieLearn course instance. Both sections, CSE 150A and CSE 250A use the Enrollment Link.
To schedule the test, first make sure you self-enroll in the correct section below.
If you are enrolled in CSE 150A, please use CSE 150A Link
If you are enrolled in CSE 250A, please use CSE 250A Link
Once you have self-enrolled in your section, follow the Registration Instructions at TTC-AP&M.
More information about testing policies and procedures can be found on the TTC’s website. You may also email tritontesting@ucsd
STUDENTS WITH OSD APPROVED ACCOMMODATIONS: If you will be utilizing accommodations for your test, you will take it at the TTC’s Pepper Canyon Hall
location. You must schedule your test at least three days in advance through the RegisterBlast system. RegisterBlast scheduling is to be done ONLY by students with OSD-approved accommodations. Tests scheduled via RegisterBlast without accommodations will be cancelled.
Make-up Test
If needed, you may schedule a make-up test in Week 10. The make-up test is cumulative and covers all material through Week 9. If your make-up test score is higher than the lower of Test 1 or Test 2, it will replace that lower score. Taking the make-up test is optional.
Under this policy, a poor performance on an earlier test can be erased by stronger performance on the make-up test. If you score lower on the make-up test, it will not reduce your grade.
Example: If you scored 60% on Test 1 and 90% on Test 2, and then earn 85% on the make-up test, your Test 1 score is replaced with 85%.
Grading¶
We will use the following grade breakdown:
Assesment | Percentage |
---|---|
Homework | 40% |
Test 1 | 15% |
Test 2 | 15% (can increase to 20%) |
Mini-Project | 10% |
Final Project | 15% |
Participation | 5% |
Redistribution¶
Participation makes up 5% of your final grade. However, any participation points you lose will add weight to Test 2. Here are example scenarios:
Scenario 1: You answer 100% of participation questions. You get the full 5% for participation and Test 2 will be worth 15%.
Scenario 2: You answer 50% of participation questions.
You’ll keep 2.5% from participation and the remaining 2.5% will be transferred to Test 2. In this case, Test 2 would be worth 17.5% of your grade instead of 15%.
Scenario 3: You don’t answer any participation questions. Test 2 will be worth 20% of your final grade instead of 15%.
This is designed to let you make up for any missed classes!
Grading Scale¶
In a typical quarter, the make-up test serves the same role as a traditional grade curve, so no additional curve is applied. The standard grading scale above is the starting point, but after all scores are in a clustering algorithm will set the final cutoffs for each letter grade in each section (150A or 250A). These cutoffs can only move down, never up. For example, the A cutoff will never exceed 94%.
Letter Grade | Range |
---|---|
A | ≥ 94% |
A- | ≥ 90% < 94% |
B+ | ≥ 87% < 90% |
B | ≥ 84% < 87% |
B- | ≥ 80% < 84% |
C+ | ≥ 77% < 80% |
C | ≥ 74% < 77% |
C- | ≥ 70% < 74% |
D+ | ≥ 67% < 70% |
D | ≥ 64% < 67% |
D- | ≥ 61% < 64% |
F | < 61% |
A+ grades have no fixed cutoff. Instead, they are awarded to the top 5% of students in each section based on overall course grade.
Calculating your grade: When calculating your homework grades, you should weigh each assignment by the points possible on that assignment. In other words, some homeworks are worth more, and some are worth less. This is by design, since some assignments are intentionally shorter.
Comprehensive Exam for MS Students: We will follow the department requirement of a B- or higher for the comprehensive exam and S/U grades.
Pass/Not Pass for Undergraduate Students: We will follow the university grading system of C- or higher for P/NP grades.
Incomplete Grade: Sometimes, circumstances beyond a student’s control prevent them from completing a class even once they have completed the majority of the coursework at a passing level. UCSD has a process in place for you to request an Incomplete (I) if this happens to you. Here is the campus policy about the Incomplete grade and some information about it.
Course Policies¶
Accommodations for Students with Disabilities¶
We aim to create an environment in which all students can succeed in this course. We need and want to hear from you if additional accommodations would improve your experience in the course.
If you have a disability, please contact the Office for Students with Disability (OSD), which is located in University Center 202 behind Center Hall, to discuss appropriate accommodations right away. They also provide the OSD Student Portal. We will work to provide you with the accommodations you need, but you must first provide a current Authorization for Accommodation (AFA) letter issued by the OSD. You are required to present the AFA letters to the Faculty and to the OSD Liaison in the department in advance so that accommodations may be arranged. We ask that you work to organize the AFA and to let us know about it as early in the quarter as possible so that we can best support your needs. For more information, see Disability Resources at UCSD.
Academic Integrity¶
In this course we expect students to adhere to the UC San Diego Integrity of Scholarship Policy. This means that you will complete your work honestly, and with integrity, and support an environment of integrity within the class.
Some examples of specific ways this policy applies to CSE 150A/250A include:
Collaboration and use of AI is allowed during homework assignments and projects,but all writeups and code must reflect your own work (or your group’s work for the final project).
Students may not look at solutions from any assignments from previous versions of the course.
Students may not post any problems or solutions from any homework or exam to any repository or website.
No unauthorized aids may be used during tests.
Late or Missed Assignments/Missed Test Policy¶
Tests must be taken at their scheduled time and may not be made up. Beyond the 24hr grace period, late homework will not be accepted. There is no grace period for project submissions. If you have a serious illness or an emergency, please talk to the instructor about it.
Outside Tutoring¶
Individuals are not permitted to approach students to offer services of any kind in exchange for pay, including tutoring services. This is considered a solicitation for business and is strictly prohibited by University policy.
Class material and intellectual property¶
Our lectures and course materials, including videos, assignments, and similar materials, are protected by U.S. copyright law and by University policy. We are the exclusive owner of the copyright in those materials we create. We acknowledge the cumulative contributions to this course material of previous instructors, TAs, and tutors, as well as contributions to the class structure from colleagues in CSE and at UCSD.
You may take notes and make copies of course materials for your own use. You may also share those materials with another student who is enrolled in or auditing this course. You may not reproduce, distribute or display (post/upload) lecture notes or recordings or course materials in any other way — whether or not a fee is charged — without our express prior written consent. You also may not allow others to do so. If you do so, you may be subject to student conduct proceedings under the UC San Diego Student Code of Conduct.
Similarly, you own the copyright in your original work. If I am interested in posting your answers or papers on the course web site, I will ask for your written permission.
Late Adds¶
I follow the CSE department guidance on Late Adds, namely that “all students are expected to attend class for the first two weeks and complete assignments if they are on the waitlist for a course. Attending class and completing course assignments does not guarantee enrollment. If students choose to miss class or not turn in assignments while on the waitlist, the student will receive a “0” on all missed assignments, if they secure a seat in the course off the waitlist.”
Exceptions and Flexibility¶
The policies in this syllabus are meant to be clear and fair, but we recognize that life does not always follow a perfect schedule. In general, our policies serve three purposes:
Support learning. Deadlines, test structures, and regrade windows help you keep pace with the material and give you timely feedback.
Provide built-in flexibility. Grace periods and make-up options (such as the 24-hour late window and the optional make-up test) are there so that an occasional conflict or busy week does not derail your progress.
Keep the course manageable. With hundreds of students, consistent rules and firm deadlines allow us to grade promptly and provide the level of support you expect.
These policies are designed to handle most of the everyday disruptions of a quarter so you do not need to request one-off exceptions for routine issues.
When to reach out Some situations go beyond these built-in allowances. Please contact the instructor as soon as possible if you experience a serious illness, hospitalization, family emergency, accident, or other significant event that prevents you from meeting course requirements. We will work with you to find an appropriate solution.
Resources for Students¶
Mental Health Services¶
For students seeking services for mental health issues (including, but not limited to: stress, sleep issues, depression, anxiety, academic distress, relationship issues, etc.), Counseling and Psychological Services (CAPS) provides free, confidential psychological counseling and crisis services for all registered UC San Diego students. CAPS also provides a variety of groups, workshops and drop-in forums.
To contact CAPS, call (858) 534-3755. All students are screened with a brief telephone assessment. For more information, visit the Counseling and Psychological Services website.
Getting Help¶
We expect that all students will need help at some point during the quarter. Many resources are available including TA and instructor office hours, as well as the Piazza discussion board. We also encourage you to form study groups but make sure that you adhere to the collaboration policy listed above in this syllabus.
The IDEA Engineering Student Center, located just off the lobby of Jacobs Hall, is a hub for student engagement, academic enrichment, personal/professional development, leadership, community involvement, and a respectful learning environment for all. The Center offers a variety of programs, listed on the IDEA Center Facebook page at http://
Diversity and Inclusion¶
We are committed to fostering a learning environment for this course that supports a diversity of thoughts, perspectives, and experiences, and respects your identities (including race, ethnicity, heritage, gender, sex, class, sexuality, religion, ability, age, educational background, etc.). Our goal is to create a diverse and inclusive learning environment where all students feel comfortable and can thrive.
Our instructional staff will make a concerted effort to be welcoming and inclusive to the wide diversity of students in this course. If there is a way we can make you feel more included please let one of the course staff know, either in person, via email/discussion board, or even in a note under the door. Our learning about diverse perspectives and identities is an ongoing process, and we welcome your perspectives and input.
We also expect that you, as a student in this course, will honor and respect your classmates, abiding by the UCSD Principles of Community (https://
If you experience any sort of harassment or discrimination, please contact the instructor as soon as possible. If you prefer to speak with someone outside of the course, please contact the Office of Prevention of Harassment and Discrimination: https://
Basic Needs/Food Insecurities¶
If you are experiencing any basic needs insecurities (food, housing, financial resources), there are resources available on campus to help, including The Hub and the Triton Food Pantry. Please visit http://
FAQs¶
Is this class curved?
Sort of. We start with the standard grading scale (A ≥ 94%, A- ≥ 90%, etc.). After all scores are in, we run a clustering algorithm to find the best cutoffs for each letter grade. These cutoffs can only be lowered—never raised—so the A threshold will never exceed 94%.
Can I take the tests remotely or schedule my test in another week?
Tests are in person at the Triton Testing Center. The self-scheduled tests already provide flexibility. Anything beyond this would require significant extra work for staff and is not allowed. Instead, the optional make-up test in Week 10 serves as a built-in backup: if you miss or underperform on an earlier test, the make-up can replace your lower score.
What if I’m mildly sick and can’t submit homework on time?
Minor illnesses are common. The 24-hour late window, optional make-up test, and participation drop are designed to absorb routine disruptions without individual exceptions. For serious illness or an extended medical issue, contact the instructor and provide documentation so we can coordinate accommodations.
I’m close to the next grade up. Can you round my final grade or give extra credit work?
No. Grades are determined by the published policies for everyone. Offering extra work or rounding for individuals would be unfair and would create grading overload for the staff. We also can’t offer special opportunities (such as extra work) to raise an individual grade. To be fair, any extra assignment would need to be offered to everyone, and even a percentage of takers would create more grading than the staff can handle at the end of the quarter. Instead, second chances are built into the course through the make-up test and other policies, which give all students the same opportunity to demonstrate mastery.