MLPR 2018
| Notes
| Lectures
| Forum
| Tutorials
| Assignments
| FAQ
| Feedback
MLPR lectures, Autumn 2018
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. I give direct links to the relevant
parts. Also, please still take your own notes in class!
- Lecture 1, Monday 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, Monday 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, Monday 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
Gradients, linear regression to logistic regression.
w3b,
w3c
- Lecture 10, Monday week 4:
pdf
Stochastic gradients, softmax regression
w3b,
w4a
- Lecture 11, Wednesday week 4:
pdf
Robust logistic regression (cost functions from probabilistic models)
w4a
- Lecture 12, Thursday week 4:
pdf
Feedforward neural nets
w4b
- Lecture 13, Monday week 5:
pdf
Fitting neural nets, initial outline of back-propagation
w4c,
w5a
- Lecture 14, Wednesday week 5:
pdf
Reverse mode differentiation (back-propagation) with matrices
w5a
- Lecture 15, Thursday week 5:
pdf
Autoencoders and PCA/linear-autoencoder demos (PCA is covered in the next lecture)
w5b
- Lecture 16, Monday week 6:
pdf
PCA continued, SVD, Netflix prize
w5b,
w6a
- Lecture 17, Wednesday week 6:
pdf
Netflix and privacy. Start of probabilistic and Bayesian regression
w6a,
w6b
- Lecture 18, Thursday week 6:
pdf
Bayesian inference and prediction
w6b,
w6c
- Lecture 19, Monday week 7:
pdf
Bayesian inference and prediction continued
w6b,
w6c
- Lecture 20, Wednesday week 7:
pdf
Bayesian linear regression review, and Bayesian model choice
w7a
Matlab/Octave or
Python demo to match end of lecture,
- Lecture 21, Thursday week 7:
pdf
Gaussian process priors
w7b,
A minimal GP demo: matlab/octave, python
Alternative GP demo: matlab/octave, python
- Lecture 22, Monday week 8:
pdf
Gaussian processes for regression and relationship to linear regression
w7b, code as above, and
w8a
- Lecture 23, Wednesday week 8:
pdf
Finish GPs (with kernel logistic regression aside)
GP readings as in Lecture 22
w8a
- Lecture 24, Thursday week 8:
pdf (there's a spurious 1/S in the final line of p5)
Bayesian logistic regression and the Laplace approximation
w8b,
w8c
(some parts will be covered next week)
- Lecture 25, Monday week 9:
pdf
More on Bayesian logistic regression and the Laplace approximation.
w8b,
w8c
- Lecture 26, Wednesday week 9:
pdf
KL-divergence and variational methods
w9a
- Lecture 27, Thursday week 9:
pdf
Some motivation for Gaussian mixture models:
w9c
Stochastic variational inference:
w9b
A minimal stochastic variational inference demo: Matlab/Octave: single-file, more complete tar-ball; Python version.
Apologies: I was forced to present the material out of order because the
document camera wasn't working (again) at the start of the lecture. I have
made a plea for the setup to be replaced.
- Lecture 28, Monday week 10:
pdf
Fitting Gaussian mixtures with gradients, or EM a bound-based optimizer
w9c,
- Lecture 29, Wednesday week 10:
pdf
More on optimization, and ensembles
w10a,
w10b.
w10c.
- 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.
