2021 - 2022 Course Highlights | MPhil in Machine Learning and Machine Intelligence

skip to content

ci_101_754x308.png

MPhil MLMI Class of 2021-2022

Dissertations

Machine Learning:

Attention-Based Sheaf Neural Networks

Autoregressive Conditional Neural Processes

Better Encoders for Neural Process Family Models

Beyond Independent Masking in Tabular Self-Supervision

Deep Reinforcement Learning for 3D Molecular Design

Domain Generalization for Robust Model-Based Offline Reinforcement Learning

Global Inducing Point Variational Approximations for Federated Bayesian Neural Networks

Graph Representation Learning for Child Mental Health Prediction

Improving Machine Learning Systems by Eliciting and Incorporating Additional Human Knowledge

Non-Gaussian Lévy Processes in Machine Learning

Outlier Detection with Hierarchical VAEs and Hamiltonian Monte Carlo

Scalable Bayesian Inference for Probabilistic Spectrotemporal Models of Ia Supernovae

Stochastic Memory for Sequence Models

Speech and Language Processing:

Automating Counterspeech in Dialogue Systems

Building a Conversational User Simulator using GANs

Effectiveness of SSL Representations for Source Separation

Joint Learning of Practical Dialogue Systems and User Simulators

Multimodal Coreference Resolution

Vision Encoders in Visual Question Answering

Computer Vision:

Understanding and Fixing the Modality Gap in Vision-Language Models

Posters from the Advanced Machine Learning Module

Conditional Neural Processes

Few-Shot Learning with Novel Metrics

First-Order Approximations for Efficient Meta-Learning

Importance Weighted Auto Encoders

Meta-Learning Approaches for Regression, Classification and Graph Representation Learning

Towards a Neural Statistician

Variational Continual Learning

Weight Uncertainty in Neural Networks