Adelaide Research & Scholarship
Adelaide Research & Scholarship (AR&S) is the University of Adelaide’s digital repository. AR&S provides a platform for the collection, organisation, access and preservation of the research and scholarly outputs of the University community in digital formats, as well as digital management of information in physical formats.
University of Adelaide higher degree by research theses are deposited into the AR&S Theses community as part of the final thesis lodgement process.
AR&S also serves as the home of the digital collections of University Library Archives and Special Collections. Items include digitized representations of physical items, such as photographs and full texts, and digital-born materials, allowing worldwide access to our heritage and research collections.
Are you a University of Adelaide researcher who would like your publications in AR&S? See our support page.
Contact us. Please email Library Discovery.
Communities in Adelaide Research & Scholarship
Select a community to browse its collections.
Recent Submissions
Glocal Energy-based Learning for Few-Shot Open-Set Recognition
(IEEE, 2023) Wang, H.; Pang, G.; Wang, P.; Zhang, L.; Wei, W.; Zhang, Y.; IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (17 Jun 2023 - 24 Jun 2023 : Vancouver, Canada)
Few-shot open-set recognition (FSOR) is a challenging task of great practical value. It aims to categorize a sample to one of the pre-defined, closed-set classes illustrated by few examples while being able to reject the sample from unknown classes. In this work, we approach the FSOR task by proposing a novel energy-based hybrid model. The model is composed of two branches, where a classification branch learns a metric to classify a sample to one of closedset classes and the energy branch explicitly estimates the open-set probability. To achieve holistic detection of openset samples, our model leverages both class-wise and pixelwise features to learn a glocal energy-based score, in which a global energy score is learned using the class-wise features, while a local energy score is learned using the pixelwise features. The model is enforced to assign large energy scores to samples that are deviated from the few-shot examples in either the class-wise features or the pixel-wise features, and to assign small energy scores otherwise. Experiments on three standard FSOR datasets show the superior performance of our model.¹
Graph-Level Anomaly Detection via Hierarchical Memory Networks
(Springer Nature, 2023) Niu, C.; Pang, G.; Chen, L.; European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) (18 Sep 2023 - 22 Sep 2023 : Turin, Italy)
Graph-level anomaly detection aims to identify abnormal graphs that exhibit deviant structures and node attributes compared to the majority in a graph set. One primary challenge is to learn normal patterns manifested in both fine-grained and holistic views of graphs for identifying graphs that are abnormal in part or in whole. To tackle this challenge, we propose a novel approach called Hierarchical Memory Networks (HimNet), which learns hierarchical memory modules—node and graph memory modules—via a graph autoencoder network architecture. The node-level memory module is trained to model fine-grained, internal graph interactions among nodes for detecting locally abnormal graphs, while the graph-level memory module is dedicated to the learning of holistic normal patterns for detecting globally abnormal graphs. The two modules are jointly optimized to detect both locally- and globally-anomalous graphs. Extensive empirical results on 16 real-world graph datasets from various domains show that i) HimNet significantly outperforms the state-of-art methods and ii) it is robust to anomaly contamination. Codes are available at: https://github.com/Niuchx/HimNet.
Health care in the metaverse
(Wiley, 2023) Pietris, J.; Tan, Y.; Chan, W.O.
High-Gain, Low-Profile, Integrable Planar Lens Antenna at 275 GHz
(IEEE, 2024) Li, M.S.; Ako, R.T.; Sriram, S.; Fumeaux, C.; Withayachumnankul, W.; IEEE 11th Asia-Pacific Conference on Antennas and Propagation (APCAP) (19 Nov 2023 - 23 Nov 2023 : Guangzhou, China)
Terahertz communications require compact, high-gain antennas that can be integrated with sources, as alternative to traditional lenses, which have bulky sizes and lack design flexibility. This paper introduces a high-gain, low-profile silicon cavity antenna integrated with a WR-3 waveguide feed. The antenna consists of a 2-mm thick silicon wafer with non-uniform hole arrays designed to locally control the effective permittivity. The holes are laser-etched on both sides of the silicon wafer, to overcome phase coverage limitations caused by the tapered side walls of the air hole during fabrication. Experimental results confirm a gain of 19 dBi at 275 GHz, with a 3-dB bandwidth of approximately 29%. We further demonstrate that the high-gain antenna can be integrated with a quarter-wave plate for circularly polarized (CP) radiation.
Indoor environmental quality and occupants' satisfaction in highrise mixed-use buildings: Preliminary results from a case study
(Architectural Science Association (ANZAScA), 2023) Croffi, J.; Soebarto, V.; Kroll, D.; Barrie, H.; 56th International Conference of the Architectural Science Association (ANZAScA) (29 Nov 2023 - 2 Dec 2023 : Launceston, Tasmania)
Indoor environmental quality (IEQ) can impact occupant’s health, productivity, and wellbeing, as evidenced in various building performance studies with an occupant-centric approach. Quantifying the impact of each of the IEQ parameters – thermal comfort, visual comfort, acoustic comfort and air quality – on occupants’ satisfaction with the indoor environment (IE) is a key step to move towards a human-centric building performance evaluation, and consequently, to improve building design and their performance in fostering IE satisfaction, enhancing occupants’ wellbeing. This study conducted a post-occupancy evaluation (POE) at U City, a high-rise mixed-use building in Adelaide, South Australia. The POE investigated the correlation between IEQ and occupants’ satisfaction with the indoor environment, collecting data from apartment residents and office workers. The data collection included surveys, monitoring of indoor environmental parameters with data loggers, observations of public space use in the building and focus groups. This paper presents the preliminary results on how each of the IEQ parameters affected the overall satisfaction with the IE as part of a broader framework to evaluate building performance in fostering wellbeing. The preliminary statistical analysis showed significant correlations between IEQ parameters and IE satisfaction for both residents and workers. Indoor temperatures, noise issues or sound quality and air quality were found to be the most significant factors that affected IE satisfaction. The outcome of this investigation will inform the elaboration of a metric that will be embedded in a building design evaluation tool based on computational 3D model analysis.