Course Syllabus

Machine Learning for Sustainable Systems

Welcome to Machine Learning for Sustainable Systems (1.C01/1.C51) in Spring 2025!

Building on core material in 6.C01/C.51, emphasizes the design and operation of sustainable systems. Illustrates how to leverage heterogeneous data from urban services, cities, and the environment, and apply machine learning methods to evaluate and/or improve sustainability solutions. Provides case studies from various domains, such as transportation and urban mobility, energy and water resources, environmental monitoring, infrastructure sensing and control, climate adaptation, and disaster resilience. Projects focus on using machine learning to identify new insights or decisions that can help engineer sustainability in societal-scale systems. Students taking graduate version complete additional assignments.

Pre-requisites: Basic programming skills. Some background in probability and linear algebra.

---

Instructors & Course Staff

Instructor: Saurabh Amin

TAs: Maxime Bouscary, Prem Talwai, William Zhang

Piazza is the preferred place to ask questions related to course assignments or grades. Posting questions on Piazza is likely to yield faster responses than sending us emails (although you are welcome to send us emails as well...)

---

Lectures & Recitations

[Lecture] Tuesday 10:30am-12pm @ 1-390

[Lecture] Thursday 10:30am-12pm @ 1-390

[Recitation] Friday 1-2pm @ 1-390

---

TA Office Hours

Wednesdays 2:30pm-3:30pm in 32-D707 (the LIDS 7th floor conference room in Stata)

Instructor Office Hours

Mondays 1-2pm @ 32D-562

1-1 meetings over zoom can be arranged upon request

---

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.

Piazza link: https://piazza.com/mit/spring2025/1c01/home 

---

Course Logistics

This course is offered as part of multi-department collaboration under the umbrella of “Common Ground for Computing Education” in the MIT Schwarzman College of Computing. This 6 unit course is to be taken jointly with another 6-unit core course 6.C01/C51 Modeling with Machine Learning: From Algorithms to Applications taught by Prof. Regina Barzilay & Prof. Marzyeh Ghassemi from MIT EECS. The core lectures focus on introducing students to modern machine learning methods, from supervised to unsupervised models, including newer neural approaches. In 1.C01/C51, we will focus on applications of core ML models covered in 6.C01/C51. We will match the ML concepts covered in core lectures and focus on modeling applications.

The emphasis is not only on learning the basic principles but also on understanding of how and why the methods work, when they are applicable, how they can be adapted and extended to solve domain-specific problems, and how the methods can and should be evaluated.

Please note the following about 1.C01/C51 logistics:

• The lectures extend the core knowledge to emphasize the applications of machine learning in the domains of climate science, energy systems, transportation and urban mobility, environmental monitoring, infrastructure sensing and control, climate adaptation, and disaster resilience. Note: unlike core lectures, we won't be able to provide recordings of 1.C01/C51 lectures
• Problem sets and exercises aim to teach how one can use heterogeneous multi-source data and apply modern machine learning methods to evaluate and improve sustainability solutions.
• The course staff will mentor student teams on their class projects that use machine learning to identify new insights or decisions to better understand/predict environmental phenomena , and improve sustainability  and resilience of societal-scale infrastructure and services.
• In place of traditional recitations, the teaching staff will hold informal (optional) practice sessions (Friday 1-2pm in 1-390) to clarify concepts and work through examples from selected topics. We encourage students to come with questions about the lecture material, recitation recordings, and problem sets. Few of these sessions will be reserved for hosting guest speakers. 
• Additionally, we will post weekly recorded demonstrations following the in-person 1.C01/C51 lecture. These recordings will focus on gaining a better understanding of the modeling approaches discussed in class by applying them towards topics related to sustainable systems. 

Grading scheme 

• Homework: 60% (4 problem sets)
  • Pset 1 – Due April 10 (Thursday)
  • Pset 2 – Due April 18 (Friday)
  • Pset 3 – Due May 1 (Thursday)
  • Pset 4 – Due May 13 (Tuesday)
• Project: 40%

Project milestones: Please follow the milestones below. All submissions should be made via Canvas.

• April 10 (Thursday). Identify project topic, modeling question, and ML models. [0%]
• April 18 (Friday). Final description of the problem, links to literature, preliminary results/insights. [7%]
• May 1 (Thursday). Progress update, 5-6 pages revisited report with descriptions of the method and preliminary results. [8%]
• May 9 (Friday) & 13 (Tuesday). In-class presentations. [10%, 3-5 mins] 
• May 15 (Thursday). Final Report and associated code. [15%]

Collaboration: We encourage students to discuss assignments with other students and with the teaching staff to better understand the concepts. When you submit an assignment under your name, however, you are certifying that the details are entirely your own work and that you played at least a substantial role in the conception stage. You should never use results (solutions, code) from other students from this year or previous years in preparing your solutions for any assignment unless you developed the materials while working with other students in class. Students should never share solutions (or staff solutions) with other students.

Extensions: Problem sets are due at 11:59pm (EST). Each day that a homework is late, it will encounter a 10% penalty, up to 30%. If an assignment is submitted beyond 3 days late, it will receive a 0. For example, if a homework which is due on Thursday warrants a 92% but was submitted 2 days late on Saturday, it will receive a 72%. If it were submitted 4 days late on Monday, it will receive a 0%.

The only exception to this is if the delay comes with an S^3 or GradSupport Dean’s support. A student with a Dean's support can submit an assignment without penalty up until the solutions are released. In the case where the situation grants an extension beyond the release of the solutions, the portion of the grade from that assignment will be redistributed among the two exams.

Guidelines on Free Expression: We refer students to the following guidelines on free expression at the Institute: https://resources.mit.edu/freeexpression-event-guidelines/

---

Course Schedule

 

Date	Session	Type	Topic	Homeworks	Project
Apr 1 (Tue)	L1	Lecture	Machine Learning for Sustainability
Apr 3 (Thu)	L2	Lecture	Data and Feature Engineering
Apr 4 (Fri)	R1	Recitation
Apr 8 (Tue)	L3	Lecture	Nonlinear Models & Ensemble Methods
Apr 10 (Thu)	L4	Lecture	Unsupervised Learning	Problem Set 1 due (15%)	Project topic due (0%)
Apr 11 (Fri)	R2	Recitation
Apr 15 (Tue)	L5	Lecture	Models for Spatiotemporal and Structured Data
Apr 17 (Thu)	L6	Lecture	ML + Optimization for Decision-Making
Apr 18 (Fri)	R3	Recitation		Problem Set 2 due (15%)	Final description due (7%)
Apr 22 (Tue)	L7	Lecture	Causality
Apr 24 (Thu)	L8	Lecture	Generative Models
Apr 25 (Fri)	R4	Recitation
Apr 29 (Tue)	L9	Lecture	Uncertainty Quantification
May 1 (Thu)	L10	Lecture	Interpretability and Fairness	Problem Set 3 due (15%)	Progress update due (8%)
May 2 (Fri)	R5	Recitation
May 6 (Tue)	L11	Lecture	Reinforcement Learning
May 8 (Thu)	L12	Lecture	Hybrid Models
May 9 (Fri)
May 13 (Tue)		Presentation		Problem Set 4 due (15%)	Presentation (10%)
May 15 (Thu)					Final report due (15%)