Welcome to 6.C01/6.C51

Modeling with Machine Learning: from Algorithms to Applications

Spring 2025

The class focuses on modeling with machine learning methods with an eye towards applications in engineering and sciences. Introduction to modern machine learning methods, from supervised to unsupervised models, with an emphasis on newer neural approaches. Emphasis on the understanding of how and why the methods work from the point of view of modeling, and when they are applicable. Using concrete examples covers the formulation of machine learning tasks, adapting and extending methods to given problems, and how the methods can and should be evaluated. Students cannot receive credit without simultaneous completion of a 6-unit disciplinary module offered by various departments in the School of Engineering.

The disciplinary modules are -

Graduate (paired with 6.C51)
1.C51, 2.C51, 3.C51, 6.C511, 7.C51, 10.C51, 20.C51, 22.C51, or SCM.C51

Undergraduate (paired with 6.C01)
1.C01, 2.C01, 3.C01, 6.C011, 7.C01, 10.C01, 20.C01, 22.C01

Instructors & Course Staff

Instructors: Regina Barzilay & Marzyeh Ghassemi

TAs: Felix Faltings, Hannes Staerk, Athena Wang, Sabrina Queipo

Instructor + TA contact: Ask a private question on Piazza.

All communication with course staff should go through Piazza rather than email.

Location: MW 2:30-4 @ 32-123

Office Hours

Monday 10-11 AM: Athena 24-323
Tuesday 9-10 AM: Hannes 32-G451
Wednesday 9:30-10:30 AM: Hannes 32-G451
Wednesday 4-5 PM: Athena 26-322
Thursday 3-4 PM: Felix 26-322
Thursday 4-5 PM: Sabrina 26-322
Friday 11 AM-12 PM: Sabrina 24-323
Friday 3:30-4:30 PM: Felix 26-322

Prerequisites and Other Resources

We recommend some chapters of the "Mathematics for ML" book: https://mml-book.github.io/book/mml-book.pdf

Linear Algebra: Sections 2.2 - 2.7

Calculus: Sections 5.1 - 5.3

Basic Probability: 6.2 and 6.3

Other resources:

MIT’s The Missing Semester A great run-through of important computing tools and basics
Weights & Biases A phenomenal site that helps you keep track of many aspects of your ML experiments
Understanding how Python handles functions
A jupyter notebook with some python basics: Python_Clinic_2168_Learning_Machines.ipynb
Outdated lecture notes, only an additional reference: lecture_notes_2016 (1).pdf

Textbook

TBD

Tentative Schedule

The class meets on Mondays and Wednesdays for a 1.5-hour lecture.

Below is a detailed lecture and quiz schedule for this semester. This syllabus focuses on the common core components of the class.

Date	Lecture #	Topic	Resources
2/3/2025	W1 L1	Introduction to Modeling: linear models	Slides, Recording
2/5/2025	W1 L2	Linear models continued, feature design	Animations
2/10/2025	W2 L3	Nonlinear prediction; neural networks	simpsons_paradox_figures.pdf
2/10/2025		Homework 1 due at 12pm (noon)	Pset
2/12/2025	W2 L4	ML packages	ml_packages_simpson.ipynb ml_packages-1.ipynb
2/17/2025		President's Day holiday
2/18/2025	W3 L5	Gradient-based learning of complex models	Slides Backprop Worked Example Gentle Explanation of Backpropagation by Andrew Glassner
2/19/2025	W3 L6	NNs for images and graphs: CNNs, GNNs	CNN.pdf Tutorial5_property_prediction_tutorial.ipynb Tutorial2_deeplearning_intro.ipynb
2/24/2025	W4 L7	Generative models: Probability, Sampling & Estimation	Slides
2/24/2025		Homework 2 due at 12pm (noon)
2/26/2025	W4 L8	No class: Quiz 1	Solutions
3/3/2025	W5 L9	Generative models: Variational Autoencoders	Slides Blog on VAEs (with code!)
3/5/2025	W5 L10	Generative models: Transformers and Foundation models	HuggingFace’s high-level summary of transformer models The Illustrated Transformer The authors of Transformers presentation of the paper Weights and Biases’ Transformer code walkthrough
3/7/2025		Homework 3 due at 5pm
3/10/2025	W6 L11	Generative models: Diffusion Models	Slides Diffusion models from scratch. Diffusion models from the discrete-time perspective (reverse a Markov process). Diffusion models from the continuous time perspective (reverse a stochastic differential equation). Annotated diffusion model pytorch implementation.
3/12/2025	W7 L12	Interpretability, bias, and robustness	Interpretability2025.pdf Relevant papers: https://arxiv.org/pdf/1602.04938v1 https://ieeexplore.ieee.org/document/8237336
3/17/2025	W8 L13	Reinforcement learning
3/17/2025		Homework 4 due at 12pm (noon)
3/19/2025	W8 L14	No class: Quiz 2

Assignments & Grading

Your final grade in the class is computed using the following breakdowns. The graduate version of the class (6.C51) involves an additional project managed within the subject-specific class.

30% Homework assignments
35% Quiz 1
35% Quiz 2

Extensions

If you need an extension for Homeworks or for a Quiz, you have to request it through https://studentlife.mit.edu/s3 or equivalent organizations for graduate students. With approved late days, you have to inform us about them in a private message on Piazza to all TAs.

The maximum possible extension duration after a homework due date are 3 days. If no homework is submitted by then, the percentage of your grade that would have been determined by this homework will instead be determined by your performance in the next Quiz.

Piazza

This term we will be using Piazza for class discussion. The system is highly catered to getting you to help fast and efficiently from classmates, the TAs, and the instructors. Rather than emailing questions about assignments to the teaching staff, please post your questions only on Piazza.

Find our class signup link here

Additional Resources

A. Academic Integrity

In this course, we will hold you to the high standard of academic integrity expected of all students at the Institute. We do this for two reasons. First, it is essential to the learning process that you are the one doing the work. We have structured the assignments in this course to enable you to gain a mastery of the course material. Failing to do the work yourself will result in a lesser understanding of the content, and therefore a less meaningful education for you. Second, it is important that there be a level playing field for all students in this course and at the Institute so that the rigor and integrity of the Institute’s educational program are maintained.

Violating the Academic Integrity policy in any way (e.g., plagiarism, unauthorized collaboration, cheating, etc.) will result in an official Institute sanction. Since a central goal of this subject is to help you become independent and critical thinkers, the use of generative AI tools to create any work in this class is prohibited. Any work submitted using AI tools will be treated as though it was plagiarized.

Possible sanctions include receiving a failing grade on the assignment or exam, being assigned a failing grade in the course, having a formal notation of disciplinary action placed on your MIT record, suspension from the Institute, and expulsion from the Institute for very serious cases.

Please review the Academic Integrity policy and related resources (e.g., working under pressure; how to paraphrase, summarize, and quote; etc.) and contact us if you have any questions about appropriate citation methods, the degree of collaboration that is permitted, or anything else related to the Academic Integrity of this course. If any part of this is confusing or uncertain, please reach out to the instructors for a conversation before submitting your work.

B. Special accommodations for students with disabilities

If you need disability-related accommodations, we encourage you to meet with me early in the semester. If you have not yet been approved for accommodations, please contact Student Disability Services at sds-all@mit.edu.

We look forward to working with you to assist you with your approved accommodations.

C. Student mental health

As a student, you may experience a range of challenges that can interfere with learning, such as strained relationships, increased anxiety, substance use, feeling down, difficulty concentrating and/or lack of motivation. These mental health concerns or stressful events may impact your ability to attend class, concentrate, complete work, take an exam, or participate in daily activities. If you are dealing with a personal or medical issue that is impacting your ability to attend class, complete work, or take an exam, you should contact a dean in Student Support Services (S3). S3 is an office that is here to help you. The deans will verify your situation, provide you with support, and help you work with your professor and/or instructor to determine next steps. In most circumstances, undergraduate students will not be excused from coursework without verification from a dean. Please visit the S3 website for contact information and more ways that they can provide support. For urgent or after-hours concerns, please contact MIT Police.

D. Diversity, equity, inclusion, and justice

Fulfilling the promise of education requires honoring individuals of all backgrounds, beliefs, ethnicities, national origins, gender identities, sexual orientations, religious and political affiliations – and other visible and nonvisible differences. All members of this class are expected to contribute to a respectful, welcoming, and inclusive environment for every other member of the class. Please feel free to speak with the teaching team if you are concerned about anything. We are accountable for the culture of this classroom, we value listening and learning, and we want to hear from you. Please also feel free to reach out to the EECS committee on DE&I at dei@eecs.mit.edu.

E. Navigating Resources Flow Chart

Here is a flow chart developed by the Undergraduate Association Committee on COVID-19 that may help you navigate academic challenges you face: https://app.lucidchart.com/documents/view/2a1cd8f8-ce7e-4398-b4b5-655c81cb54cc/0_0Links to an external site.