Skip to content

Spring 2025 Overview: HW1–7 with Code and Data You Can Run, Plus What Fall 2026 Looks Like | Berkeley CS189 Course Overview

Aug 22, 20261 min
Home›learning
TL;DRCS189 exists online in several terms. From post 2 onward this series uses Spring 2026 (eecs189.org/sp26, A3) as its base, walking through each lecture block and assignment; Spring 2025 (Shewchuk) is a different classic route with 25 lectures of notes, HW1–7 and past exams, but its official recordings sit behind a bCourses login; Fall 2026 is still in progress and serves only as a reference. This post is the series entry point and full table of contents.
#berkeley#cs189#machine-learning#open-course#learning-path
Series: Reading Berkeley CS189 Introduction to Machine Learning (1 / 18)
Three Versions of CS189: Spring 2026 as the Base, Spring 2025 as the Classic, Fall 2026 in Progress | Berkeley CS189 Course Map

🌏 中文版

Video status: Official entry or recording index only. Source details

Berkeley CS189 Introduction to Machine Learning is the math-heavy ML entry point at Berkeley. It does not follow CS188 Spring 2026; the two are parallel. The Berkeley AI/ML Course Guide frames them as search/reasoning vs. statistical learning — this post makes CS189 executable.

To self-study lecture by lecture with this series, use Spring 2026; for the classic SVM / decision tree / PCA / boosting route, use Spring 2025; to see what next term looks like, watch Fall 2026. Spring 2026 publishes lecture notes, 25 lecture videos, discussions with solutions, and the HW1–5 handouts, and this series uses it as the base from post 2 onward (see Three Versions of CS189). Spring 2025 at people.eecs.berkeley.edu/~jrs/189s25/ keeps 25 lectures of notes, HW1–7, code/data and past exams (A3 in this site's A0–A3 scale); its official recordings are on bCourses and require a login. Fall 2026 at eecs189.org/fa26 just published a 27-lecture calendar (Lec01 Introduction + ML Problem Framing to Lec27 Closing), but most decks and assignments are still TBD and the rotating eecs189.org domain 302s to the current term, so old URLs can 404.

How public is it

EditionLevelWhat an anonymous reader getsMain gap
CS189 Spring 2026 (eecs189.org/sp26)A3 (series base)27-lecture schedule, lecture PDFs/notes, 25 lecture videos linked from the official site, Discussions 1–12 with solutions and walkthroughs, HW1–5 handouts (HW5 with solutions), midterm with solutionsGradescope, hidden tests, Ed; HW1–4 solutions; final exam; Lec 1 recording
CS189 Spring 2025 (~jrs/189s25)A3 (different route)25 lectures of notes, the full machlearn.pdf, HW1–7, code/data (HW4 data on Kaggle), past exams, full syllabusofficial recordings on bCourses need a login, only screen-only backup screencasts are public; Gradescope, Ed, hidden tests
CS189 Fall 2026 (eecs189.org/fa26)A1→A2syllabus, 27-lecture calendar (Week 1–16)most decks/videos/HW starters not yet released

Both A3 editions earn the grade because the practice loop is closed: problem set + runnable code/data + past exams for self-checking. Fall 2026 is useful to see the arc — from the Fall 2026 Schedule: Data Tools / K-Means / KNN → Density Estimation / GMM → Linear Regression / Bias-Variance → Logistic Regression → Gradient Descent → Neural Networks → CNN / Transformers / LLM → Attention / MDP / RL → Post-training / Diffusion → Closing — consistent with the canonical ML spine, but not a releasable self-study bundle yet.

Prerequisites

Official: multivariable calculus, linear algebra, and CS70 (or consent). An off-campus checklist:

  1. Linear algebra: matrix products, eigenvalues, SVD geometry; read the normal equations for least squares and ridge.
  2. Probability: conditional probability, expectation, MLE/MAP, bias-variance.
  3. Implementation: Python + NumPy for vectorized code and gradient checks. If shaky, shore up with CS61B Fall 2025 habits for data structures and testing.

CS189 does not list CS61B by number, but HW code assumes reproducible experiments, train/validation splits, and learning curves. Missing that hurts more than a missing course code.

The Spring 2025 lecture sequence and HW1–7

Spring 2025 is Jonathan Shewchuk's edition, and its sequence differs from Spring 2026 and Fall 2026. Going by the lecture titles on the 189s25 course page, the 25 lectures fall into five blocks:

  1. Linear classifiers (Lec 1–5): classification and train/validation/test, perceptrons, gradient descent and SGD, hard- and soft-margin SVMs, ML abstractions and types of optimization problems.
  2. Decision theory and Gaussian models (Lec 6–9): the Bayes decision rule, generative vs. discriminative models, GDA (QDA/LDA) and MLE, eigendecomposition and anisotropic Gaussians.
  3. Regression (Lec 10–13): least squares, logistic regression with Newton's method, ROC curves, statistical justifications for regression, ridge and MAP.
  4. Trees and neural networks (Lec 14–19, 22–23): decision trees, bagging and random forests, backpropagation, vanishing gradients and ReLUs, CNNs, then data augmentation, regularization, batch normalization, ResNets and AdamW.
  5. Unsupervised learning and nearest neighbors (Lec 20–21, 24–25): PCA, SVD, k-means and hierarchical clustering, AdaBoost, kNN and k-d trees.

HW1–7 were due 1/29, 2/12, 2/26, 3/12, 4/3, 4/25 and 5/7. The course page lists each assignment's files and due date but does not label topics, so this post does not assign topic tags to each HW; open the written-part PDF to see what each one covers. Every HW ships problems plus locally runnable code/data. Off-campus readers miss the hidden Gradescope tests and staff feedback, so use past exams and your own validation curves instead.

SVMs, decision trees/random forests, PCA/SVD, boosting and k-d trees are covered only in Spring 2025. When the Spring 2026-based posts reach those topics, they point back here.

Why not chase Fall 2026 directly

eecs189.org is a rotating site already 302ing to /fa26; prior term URLs can break after the switch (the Berkeley guide documents 404s). Fall 2026 currently shows a full calendar with many TBD decks/HWs. Build the spine on Spring 2026, which has finished and has its materials in place (posts 2–18 of this series), use Fall 2026's 27-lecture order as a reference, and use Spring 2025 for the classic topics.

Posts in this series

Spring 2026 has 27 lectures and 5 assignments; this series interleaves lecture blocks and assignments in the official order:

  1. This post: series entry point and edition overview
  2. Three Versions of CS189: Spring 2026 as the Base, Spring 2025 as the Classic, Fall 2026 in Progress
  3. Lec 1–3: ML Problem Framing, Data Tools, Terminology and Techniques
  4. Lec 4–7: K-means, Probability Review, MLE, Multivariate Gaussians and GMMs
  5. Lec 7–10: Linear Regression, the Geometry of Least Squares, and Regularization
  6. HW1 Guide: Linear Algebra / Calculus / Probability Warm-up + Fashion Coding
  7. Lec 11–12: Classification, Generative Classifiers, Logistic Regression, and ROC
  8. Lec 13 & 15: Convergence, Momentum, Adam, SGD
  9. Lec 14 & 16: MLE vs MAP, Bias-Variance, Entropy and KL, Plus a Midterm Self-Check
  10. HW2 Guide: Chatbot Arena Paper Questions, Regression, MLE/MAP, From GMMs to Flow Matching
  11. Lec 17–18: Depth, Universal Approximation, Activations, and Backpropagation
  12. HW3 Guide: Autograd from Scratch (BearTensor), Newton's Method, and the Information Bottleneck
  13. Lec 19–20: Initialization, BatchNorm, CNNs, Early Stopping, and Double Descent
  14. Lec 21–22: Transformers
  15. HW4 Guide: ResNet/Transformer Paper Questions and Implementing CNN, ResNet, Transformer, DNABERT, and ConvNeXt
  16. Lec 23–24: LLM Training and Applications, Self-Supervised Learning
  17. Lec 25–27: AI for Protein Engineering, Agents and Environments, and Where to Go Next
  18. HW5 (Optional) Guide: InfoNCE for Biology, Diffusion Theory, LLM Fine-Tuning + Kaggle

Still unavailable, and only noted as existing in the series: the Spring 2026 final exam and its solutions (the past-exam folder stops at the sp25/fa25 finals), official HW1–4 solutions, the Lec 26 guest lecture (no public materials), and the Lec 1 recording. Once Fall 2026 ends, the series will reassess whether to switch to it as the base.

Tonight's starter

  1. Read Three Versions of CS189 and decide between the Spring 2026 main line and the Spring 2025 classic route.
  2. On Spring 2026, open the Lec 1–3 guide and label each HW1 warm-up problem as derivation, data analysis, or programming.
  3. If the linear algebra or probability warm-ups stall you, patch fundamentals via CS70 notes or CS61B before jumping ahead.

Course video sources

This post is a course overview or resource map with no single corresponding lecture. Checked live on 2026-10-10: the Spring 2026 official YouTube playlist (25 lecture videos) is publicly viewable. The Spring 2025 official lecture videos are on bCourses and redirect to a CalNet sign-in page; the only public Spring 2025 recordings are screen-only backup screencasts on Google Drive. Fall 2026 was not part of this check.

Course and recording entries:

Checked: 2026-10-10.

Update Log

  • 2026-10-10: Rechecked video status. Confirmed the Spring 2026 YouTube playlist is public and that Spring 2025 official recordings sit behind a CalNet sign-in; the status stays official entry only for this overview.

  • 2026-10-10: Added explicit video status and checked recording sources and access notes.

  • 2026-09-29: Expanded the series to 18 posts with Spring 2026 as the base; added a "Posts in this series" table of contents and a list of unreleased parts; corrected the "HW1–7 route" section, which had presented the Spring 2026/Fall 2026 sequence as Spring 2025's HW topics, rewriting it from the lecture titles on the 189s25 course page and noting that Spring 2025 recordings need a bCourses login; added Spring 2026 to the access table; pointed the starter steps at the series posts.

References

Related

Three Versions of CS189: Spring 2026 as the Base, Spring 2025 as the Classic, Fall 2026 in Progress | Berkeley CS189 Course Map1 minML Problem Framing, Data Tools, Terminology and Techniques | Berkeley CS189 Lectures 1–31 min& 15: Convergence, Momentum, Adam, SGD | Berkeley CS189 Lecture 131 min