Assignments and Grading
Introduction to Machine Learning
This course emphasizes understanding, experimentation, and responsible use of machine learning. Assessment is designed to reward thoughtful work, reproducibility, and clear communication—not memorization or superficial results.
Grading Breakdown
| Component | Percentage |
|---|---|
| Programming Assignments (4) | 30% |
| Lab Work (6) | 20% |
| Midterm Exam | 20% |
| Final Project | 25% |
| Participation & Engagement | 5% |
| Total | 100% |
Programming Assignments (30%)
Description
Students complete four structured programming assignments focused on implementing and analyzing machine learning algorithms using Python and open-source libraries.
Assignments build progressively in difficulty and reinforce lecture and lab material.
Topics May Include
- Linear and logistic regression
- Classification methods
- Model evaluation and validation
- Feature engineering
Submission Requirements
- Jupyter Notebook (
.ipynb) - Clear code comments
- Written explanations (markdown cells)
- Reproducible results (same output when rerun)
Programming Assignment Rubric (per assignment)
| Criterion | Weight |
|---|---|
| Correct implementation | 40% |
| Code clarity & organization | 20% |
| Explanation & interpretation | 20% |
| Reproducibility | 10% |
| Proper citation of sources | 10% |
Lab Work (20%)
Description
Labs are hands-on, exploratory exercises designed to reinforce concepts through guided experimentation. Labs are usually completed during or shortly after class.
Lab Expectations
- Short code exercises
- Experimentation with parameters
- Brief reflection on results
Labs emphasize learning through doing, not perfect outcomes.
Lab Rubric (each lab)
| Criterion | Weight |
|---|---|
| Completion & effort | 40% |
| Correct use of tools | 30% |
| Interpretation of results | 20% |
| Organization & clarity | 10% |
Midterm Exam (20%)
Format
- Combination of:
- Short answer questions
- Conceptual explanations
- Light coding or pseudocode
- Closed-book, but concept-focused, not formula memorization
Covers
- Weeks 1–7
- Core ML concepts
- Model intuition and evaluation
Evaluation Focus
- Conceptual understanding
- Ability to explain algorithm behavior
- Clear reasoning
Final Project & Presentation (25%)
Overview
The final project is a comprehensive, applied machine learning project using an open dataset. Students will design, implement, evaluate, and clearly communicate a machine learning workflow.
A short presentation is a required component of the final project and is graded as part of the project score.
Project Components (Required)
Each final project must include:
- Problem Statement
- Clearly defined task (classification, regression, clustering, etc.)
- Motivation and relevance
- Dataset Description & Provenance
- Source of the dataset (must be open)
- Description of features and labels
- Discussion of limitations or biases
- Model Implementation
- At least one machine learning model
- Clear explanation of design choices
- Evaluation & Interpretation
- Appropriate metrics
- Analysis of results (not just accuracy)
- Ethical Considerations
- Bias, fairness, privacy, or misuse concerns
- Reflection on societal impact
- Reproducible Notebook
- Clean, well-documented Jupyter notebook
- Results reproducible when rerun
- Final Presentation (Required)
- Short oral presentation during Week 14
Final Project Presentation
Format
- Length: 6–8 minutes per student
- Audience: Classmates and instructor
- Visuals: Slides or notebook walkthrough (PDF or HTML)
Presentation Should Cover
- Problem and dataset
- Model approach
- Key results
- One challenge or limitation
- One ethical or interpretive insight
The goal is clear communication, not technical perfection.
Final Project & Presentation Rubric (25%)
| Criterion | Weight |
|---|---|
| Problem formulation & motivation | 12% |
| Dataset choice & documentation | 12% |
| Model implementation | 20% |
| Evaluation & interpretation | 18% |
| Reproducibility & organization | 13% |
| Ethical analysis | 10% |
| Presentation clarity & communication | 15% |
| Total | 100% |
(The presentation is fully integrated into the project grade.)
Participation & Engagement (5%)
Participation includes:
- Attending class
- Contributing to discussions
- Asking thoughtful questions
- Peer feedback on projects
The quality of engagement matters more than the quantity.
Late Work Policy
- Assignments submitted up to 48 hours late incur a 10% penalty per day
- No submissions accepted after 48 hours unless prior arrangements are made
- Extensions may be granted for documented circumstances
Collaboration Policy
- Discussion and idea-sharing are encouraged
- Code must be written individually unless stated otherwise
- Any collaboration must be acknowledged
Use of AI Tools
Students may use AI-based tools for learning support, such as:
- Debugging help
- Concept clarification
However:
- Submitted work must reflect the student’s own understanding
- Any AI assistance must be disclosed
- Copying generated solutions without understanding is prohibited
Academic Integrity
All students must follow CUNY academic integrity policies.
Open resources must be cited properly, and all submitted work must be original.
Open Education & Reuse Statement
Students are encouraged to:
- Keep copies of their work
- Reuse code responsibly
- Share projects under open licenses when possible
Learning extends beyond the classroom.
Grading Scale Link: https://hanifml2026.commons.gc.cuny.edu/grading-scale/
