CMU 10-301/601 Spring 2026: Learn Machine Learning Through Nine Assignments

TL;DR Spring 2026 publishes material for 27 lectures and nine homework bundles; outsiders can do the core work but cannot access Panopto, Piazza, Gradescope, or official homework solutions.

#cmu #machine-learning #course-guide #self-study

Table of Contents

Why follow assignments
Prerequisites and tools
Nine capability checkpoints
The access boundary
Start tonight
References

Series: Reading CMU 10-301 Machine Learning (0 / 10)

CMU 10-301 HW1: Find ML Foundation Gaps with Mathematics and Python

🌏 中文版

CMU 10-301/601 Spring 2026 is one Introduction to Machine Learning class under two numbers. Its public schedule organizes 27 lectures with readings, slides, inked slides, recitations, and exam review. The coursework page publishes nine homework bundles.

Why follow assignments

Current recordings live in Panopto, so this is not an anonymous video course. Splitting it mechanically into 27 slide summaries would hide the delivery points. The assignments move from mathematical and Python foundations through trees, linear models, neural networks, learning theory, deep learning, reinforcement learning, and unsupervised methods.

Prerequisites and tools

Bring Python, NumPy, calculus, linear algebra, and elementary probability. HW1 is the honest entry test. Follow each handout's package requirements; outside CMU, use a clean local environment and retain commands and outputs for reproducibility.

Nine capability checkpoints

Background Material: mathematics, Python, and data inspection.
Decision Trees: information, recursive induction, and evaluation.
K-NN, Perceptron, Regression: compare inductive biases.
Logistic Regression: likelihood, gradients, and implementation.
Neural Networks: computation graphs and backpropagation.
Learning Theory and Ethics: generalization, MLE/MAP, fairness metrics, and social impact.
Deep Learning: extend basic networks into deeper architectures.
Reinforcement Learning: MDPs, value, and learning through interaction.
Learning Paradigms: ensembles, k-means, PCA, and recommenders.

The access boundary

The course home establishes identity and policies, the schedule publishes readings/slides/recitations, and coursework links handouts, templates, starter assets, data, and selected reference outputs. Together these support an A3 self-study route. A3 does not equal enrollment: Panopto video, Piazza discussion, Gradescope hidden tests, staff feedback, and official homework solutions remain restricted. Public practice-exam solutions do not imply public homework solutions.

Start tonight

Download HW1. Complete one page of mathematics and the data-inspection program, compare formatting against public reference output, then record whether the failure came from a formula, array shape, or command-line interface.