Week 1

Recap of probability theory: conditional probabilities, Bayes’ theorem with several examples. Statistical inference.

Week 2

Supervised learning: the perceptron. AND, OR and XOR functions. The perceptron learning rule.

Week 3

Gradient descent learning. Learning as inference.

Week 4

Multi-layer networks: XOR function; error back-propagation. Deep learning.

Week 5

Unsupervised learning: Principal Component Analysis, Oja’s rule.

Week 6

Recap of statistical mechanics: ensembles, statistical entropy, Ising model with nearest neighbor interactions.

Week 7

Auto-associative memory and attractor neural networks. Hebbian learning rule, synaptic plasticity. Memory as an Ising model: the Hopfield network. Statistical mechanics of the Hopfield network.

Week 8

Computation of the capacity of the Hopfield network using mean field theory. Absence of spurious retrieval states; phase diagram in the temperature-storage plane. The brain as an anticipating machine: Boltzmann machines, Helmholtz machines learn probability distributions. Wake and sleep learning rule.

Week 9

Backpropagation through time. Reinforcement learning.

Week 10

The temporal credit assignment problem and its solution using temporal difference learning. The actor/critic model.