-
This research aims to design a sustainable framework for optimizing distributed computing systems to enhance performance while minimizing energy consumption. Existing scheduling algorithms often
-
models Uncertainty estimation and calibration Failure and out-of-distribution detection Adaptive and selective computation Model routing and dynamic inference Generalisation to unseen environments Embodied
-
that can better utilise distributed computing resources while reducing energy overheads and supporting more sustainable AI infrastructure. The successful candidate will have the opportunity to work on real
-
hospital or population often fail when applied elsewhere due to distributional shifts. Since acquiring new labeled data is often costly or infeasible due to rare diseases, limited expert availability, and
-
paradigms rely on a fragile "closed-world" assumption: that the unlabeled pool perfectly reflects the distribution of the labelled seed set. In real-world deployments, this is rarely true. Data streams
-
management, distributed computing, and energy-aware computing, preparing them for impactful roles in industry and research. Key Components and Example Scenarios Predictive Resource Allocation and Load
-
of Machine Learning (ML) models across large-scale distributed systems. Leveraging advanced AI and distributed computing strategies, this project focuses on deploying ML models on real-world distributed
-
Background and Motivation Modern deep learning models have achieved remarkable success in computer vision and natural language processing. However, they typically produce overconfident predictions
-
complexity), Vol. 42, No. 4, pp270-283 Wallace, C.S. and D.L. Dowe (2000). MML clustering of multi-state, Poisson, von Mises circular and Gaussian distributions, Statistics and Computing, Vol. 10, No. 1, Jan
-
at one time. In non-stationary environments on the other hand, the same algorithms cannot be applied as the underlying data distributions change constantly and the same models are not valid. Hence, we need