2016-2017 Course Highlights

Posters from the Advanced Machine Learning Module

• Auto-Encoding Variational Bayes
• Sequential Neural Models with Stochastic Layers
• Stacked Convolutional Auto-encoders for Hierarchical Feature Extraction
• Importance Weighted Autoencoders
• Semi-Supervised Learning with Deep Generative Models

Dissertations and Posters presented at the Industry Day on 19 June 2017

• Tradeoffs in Neural Variational Inference * Poster
• Memory Networks for Language Modelling
• Pathologies of Deep Sparse Gaussian Process Regression
• Waveform Level Synthesis
• Wasserstein Generative Adversarial Network
• Hierarchical Dialogue Management * Poster
• Bayesian Deep Generative Models for Semi-Supervised and Active Learning * Poster
• Improved Interpretability and Generalisation for Deep Learning
• Constrained Bayesian Optimization for Automatic Chemical Design * Poster
• Designing Neural Network Hardware Accelerators Using Deep Gaussian Processes
• Neural Program Lattices: Learning through Weak Supervision * Poster
• Natural Language to Neural Programs
• Bayesian Neural Networks for K-Shot Learning * Poster
• Bayes By Backprop Neural Networks for Dialogue Management
• Improving Sample Efficiency for Gradient-Based Policy Optimisation * Poster
• Sample Efficient Deep Reinforcement Learning for Dialogue Systems with Large Action Spaces