PSI 2018/2019 - Machine Learning (Hayward)

Generative modelling: explicit and approximate likelihood models; implicit likelihood models

Generative modelling: explicit and tractable likelihood models

Quantum state reconstruction using restricted Boltzmann machines

Restricted Boltzmann machines: training to minimize the Kullback-Liebler divergence

Generative modelling: Hopfield networks, Boltzmann machines and restricted Boltzmann machines (RBMs)

Reinforcement learning: Q-learning, Bellman equations

Reinforcement learning: Markov decision processes, policy gradient methods

Dimensional reduction using t-distributed stochastic neighbour embedding (t-SNE); Kullback-Liebler (KL) divergence; maximum likelihood estimation

Introduction to unsupervised learning; dimensional reduction using principal component analysis

Convolutional neural networks: local receptive fields, shared weights and biases, pooling

Monte Carlo sampling; summary of hyperparameters in feedforward neural networks