MLPR 2019
| Notes
| Lectures
| Forum
| Tutorials
| Assignments
| FAQ
| Feedback
MLPR lectures, Autumn 2019
Recordings: The lectures will be
recorded and will
be available later the same day (usually within 2hrs after the lecture).
If Learn knows you're taking the class, make sure you are
logged in to Learn,
then you can
access the lectures.
PDF scans: The PDF links below are to scans of what was handwritten
during each lecture. If anything is unclear, please refer to the actual class notes. We give direct links to the relevant
parts. Also, please still take your own notes in class!
- Lecture 1, Tuesday week 1:
pdf
Logistics and motivation.
w0a–w0f, w1a
- Lecture 2, Wednesday week 1:
pdf
Linear functions and least squares.
w1b
- Lecture 3, Thursday week 1:
pdf
Introduction to basis functions and L2 regularization.
w1c
- Lecture 4, Tuesday week 2:
pdf
Generalization and dataset splits.
w2a
- Lecture 5, Wednesday week 2:
pdf
More generalization, Gaussians, CLT.
w2b, w2c
- Lecture 6, Thursday week 2:
pdf
Standard error bars, different sources of variability. Multivariate Gaussians.
w2d, w2e
- Lecture 7, Tuesday week 3:
pdf
Gaussian classifiers.
w3a
- Lecture 8, Wednesday week 3:
pdf
More on baseline classifiers, regressing on labels.
w3a
- Lecture 9, Thursday week 3:
pdf
Start of probabilistic and Bayesian regression.
w3b
- Lecture 10, Tuesday week 4:
pdf
Bayesian inference and prediction.
w4a
- Lecture 11, Wednesday week 4:
pdf
Bayesian inference and prediction continued.
w4a
- Lecture 12, Thursday week 4:
pdf
Bayesian model choice.
w4b
- Lecture 13, Tuesday week 5:
pdf
Gaussian process priors
w5a,
A minimal GP demo: matlab/octave, python
Alternative GP demo: matlab/octave, python
- Lecture 14, Wednesday week 5:
pdf
Gaussian processes for regression and relationship to linear regression
w5a, code as above, and
w5b
- Lecture 15, Thursday week 5:
pdf
Finish GPs
GP readings as in Lecture 14
w5b
- Lecture 16, Tuesday week 6:
pdf
Gradients, linear regression to logistic regression.
w6a,
w6b
- Lecture 17, Wednesday week 6:
pdf
Stochastic gradients, softmax regression
w3b,
w4a
- Lecture 18, Thursday week 6:
pdf
Robust logistic regression (cost functions from probabilistic models)
w6c
- Lecture 19, Tuesday week 7:
pdf
Feedforward neural nets
w7a
- Lecture 20, Wednesday week 7:
pdf
More on neural nets and their fitting
w7b
- Lecture 21, Thursday week 7:
pdf
Early stopping.
Reverse mode differentiation (back-propagation) with matrices
w7b,
w7c
- Lecture 22, Tuesday week 8:
pdf
Autoencoders and PCA/linear-autoencoder demos (more on PCA in the next lecture)
w8a
- Lecture 23, Wednesday week 8:
pdf
PCA continued, SVD, Netflix prize
w8a,
w8b
- Lecture 24, Thursday week 8:
pdf
Netflix prize and privacy. Start Bayesian logistic regression
w8b,
w8c,
w8d
- Lecture 25, Tuesday week 9:
pdf
Guest lecture by John Quinn. Notes and source code.
- Lecture 26, Wednesday week 9:
pdf
More on Bayesian logistic regression, and the Laplace approximation.
w8c
- Lecture 27, Thursday week 9:
pdf
More on Bayesian logistic regression, and the Laplace approximation.
w8c,
w8d
- Lecture 28, Tuesday week 10:
pdf
Monte Carlo prediction, KL-divergence and variational methods
w8d,
w9a
- Lecture 29, Wednesday week 10:
pdf
Stochastic variational inference:
w9b
Some motivation for Gaussian mixture models:
w9c
A minimal stochastic variational inference demo: Matlab/Octave: single-file, more complete tar-ball; Python version.
- Lecture 30, Thursday week 10:
Exam preparation advice
(alternative EASE link).
That’s all folks! Have a great winter break.
Those enrolled on the class, please take the class survey.
