4 – Applying attention mechanisms to reinforcement learning in traffic

In reinforcement learning, agents usually process all inputs equally, even if some are more relevant than others. This project will explore the use of a simplified attention mechanism in a car-following environment, allowing a vehicle to focus more on important neighbours (such as the vehicle directly in front). Students will compare the performance of Multi-Agent … Read more

3 – Tackling the cold-start problem in multi-agent reinforcement learning for cooperative driving

Reinforcement learning agents often behave conservatively or inconsistently at the start of training, leading to the so-called cold-start problem. This project will simulate a simple traffic scenario and train a multi-agent RL-based autonomous vehicle controller. Students will then explore strategies to reduce cold-start effects, such as imitation learning for initialization or reusing past experiences, and … Read more

5 – Laundry drying assistant using hyper local weather prediction

Drying clothes outdoors is one of the most affordable and environmentally friendly methods, but in Ireland’s unpredictable weather, laundry that is nearly dry can quickly become wet again. This project will focus on the design and development of a smart laundry drying assistant that helps users make better decisions about when to dry clothes outside. … Read more

1 – Cooperative CAV decision-making using Graph Neural Networks

Before fully autonomous driving is achieved, connected autonomous vehicles (CAVs) will operate for a certain period in mixed traffic, which includes both CAVs and human-driven vehicles (HDVs). The dynamic and interactive conditions in mixed traffic scenarios renders CAV decision making particularly challenging. This project will investigate the use of Graph Convolutional  Deep Reinforcement Learning for … Read more

To interpret or not to interpret: Forecasting conflict fatalities with machine learning models vs. GLMs

Forecasting conflict on a fine-grained grid level has real-life policy implications that can empirically inform meaningful healthcare and peace-preservation decisions. In many settings, interpretable models have the appeal that policymakers know how to draw conclusions from the model and do not have to base their decisions on black-box models. On the other hand, machine learning … Read more

Hyperscanning interactive dialogue (taken)

This project will investigate how individuals adapt to each other during a dialogue. Rather than focusing on speech recordings only, this project will involve the data collection and analysis of neural data from electroencephalography from two individuals simultaneously as a dialogue unfolds. This is part of a bigger project led by a PhD students in … Read more

Speechify – Let me speak again! (taken)

Restoring speech communication in individuals that can’t speak or have difficulty speaking (e.g., individuals with ALS) is an important challenge that could benefit many. While many solutions have been proposed that have various limitations, recent progress in machine learning methods (e.g., transformers) open new opportunities. This project aims to decode the intended speech of individuals … Read more

Boosting Language Learning with Machine Learning (taken)

Learning a second language is a challenging and time-consuming task. While a myriad of strategies have been proposed, we are still far from reaching an optimal personalised solution. Users either drop out early in the process or, once they reach a sufficient level of proficiency, they hit a plateau that seems impossible to overcome. This … Read more

Generative AI for Immersive 3D Experience

We will explore generative AI with a focus on 3D content creation and editing. You will build up your background in 3D computer graphics and 3D computer vision and conduct research in the following themes including neural radiance fields, diffusion models, or applied projects. 1. Neural radiance fields Project 1.1: Real-time neural radiance fields. We … Read more

2 – Deep RL-based CAV longitudinal controller in mixed traffic flow

Connected Autonomous Vehicles (CAVs) are expected to share the roads with Human Driven Vehicles for the foreseeable. The random sequences in this mixed traffic flow render the design of CAV controllers particularly challenging.  This project investigates the design of a deep reinforcement learning algorithm  algorithms to reduce the training dimensions and alleviate computational burdens. This … Read more