Kevin Wong

Explainable Artificial Intelligence methods have gained significant traction for their ability to elucidate the decision-making processes of black-box models, particularly in high-stakes fields such as healthcare and finance. Among these, Local Interpretable Model-agnostic Explanations (LIME) stands out as a widely adopted post-hoc, model-agnostic approach that interprets black-box predictions by constructing an interpretable surrogate model on perturbed instances to approximate the local behavior of the original model around a given instance. However, the effectiveness of LIME can depend on the quality of the training data used by the black-box model. When trained on limited or low-quality data, the black-box model may yield inaccurate predictions for perturbed samples, resulting in poorly defined local decision boundaries and consequently unreliable explanations. This limitation is especially problematic in data-scarce settings. To overcome this challenge, we propose TransLIME, a novel end-to-end explainable transfer learning framework that improves the local fidelity and stability of LIME on limited tabular datasets by transferring relevant explainability knowledge from a related auxiliary source domain with a shifted distribution. Also, in TransLIME, only representative source prototype explanations obtained through clustering are transferred to the target domain, thereby reducing cross-domain exposure of both data and explanatory information during transfer. Experimental evaluations on real-world datasets demonstrate the effectiveness of the proposed framework in improving explanation quality in target domains with limited data.

While Large Language Models (LLMs) have shown remarkable proficiency in text-based tasks, they struggle to interact effectively with the more realistic world without the perceptions of other modalities such as visual and audio. Multi-modal LLMs, which integrate these additional modalities, have become increasingly important across various domains. Despite the significant advancements and potential of multi-modal LLMs, there has been no comprehensive PRISMA-based systematic review that examines their applications across different domains. The objective of this work is to fill this gap by systematically reviewing and synthesising the quantitative research literature on domain-specific applications of multi-modal LLMs. This systematic review follows the PRISMA guidelines to analyse research literature published after 2022, the release of OpenAI’s ChatGPT
3.5. The literature search was conducted across several online databases, including Nature, Scopus, and Google Scholar. A total of 22 studies were identified, with 11 focusing on the medical domain, 3 on autonomous driving, and 2 on geometric analysis. The remaining studies covered a range of topics, with one each on climate, music, e-commerce, sentiment analysis, human-robot interaction, and construction. This review provides a comprehensive overview of the current state of multi-modal LLMs, highlights their domain-specific applications, and identifies gaps and future research directions.

Integration of renewable energy sources like solar and wind power into the power network has increased significantly in recent years. However, these sources are inherently variable and intermittent, which leads to challenges in maintaining grid stability and reliability. A promising solution to these challenges is the strategic deployment of battery energy storage systems (BESS). The BESS can support improving system voltage and frequency stability and increase system reliability because it can rapidly charge and discharge the grid when needed. To fully explore the advantages of BESS in power systems, it is crucial to determine their optimal allocation. Therefore, this paper presents a technique for optimal allocation of BESS in weak grids to bolster system voltage and frequency stability and enhance system reliability. The proposed method uses the recent adaptive grey wolf optimisation (AGWO) algorithm to identify the optimal capacity and placement of the BESS. The AGWO algorithm is a metaheuristic optimisation algorithm that uses a population of wolves to explore the solution space for the best outcome. The outcomes from the AGWO method are validated using grey wolf optimisation (GWO), beluga whale optimisation (BWO), and sparrow search algorithm (SSA). The efficacy of the proposed methodology is validated in a high renewable distributed generation (DG) penetrated weak IEEE-39 bus system using DIgSILENT PowerFactory software. Simulation findings demonstrate that integrating BESS at the optimal location and size can significantly improve the voltage and frequency stability of the grid and increase its reliability. The proposed methodology can help grid operators and system planners make informed decisions on integrating BESS into the grid.

This study initially undertook a scoping review of mobile Computer-Supported Collaborative Learning (mCSCL) to determine the characteristics, and methodologies of the reviewed studies, and its benefits and challenges in resource-constrained nations. Online databases yielded no related articles on mCSCL in resource-constrained countries from 2007 to 2023. Nevertheless, 32 papers centred on mobile learning (m-learning) in resource-constrained countries were identified. Findings from the research questions encompass the theoretical framework used, the study context, sample size, and research design of the reviewed articles; and the benefits, and challenges of m-learning in resource-constrained countries. Implications for future m-learning and mCSCL studies in resource-constrained countries are considered. These include focussing on the benefits of m-learning as a basis to foster its implementation in the region, while providing adequate measures to mitigate its challenges, potential areas for investigation, such as research situated in resource-constrained countries in South American and Eastern Europe and research on aspects of m-learning such as mCSCL, more strategically focussed research interventions utilising available mobile infrastructure in developing countries, studies grounded in established theoretical frameworks and seminal studies on research methodology to inform research design.

The main contribution of this paper is to introduce a framework for integrating Machine Learning (ML), Human, and Virtual Reality (VR) into one platform to promote a collaborative visualisation environment that can assist in better analysis and improve the human-machine teaming capability. This platform was demonstrated using a case study in ana-lysing road traffic conditions. The ‘Ab-normal Machine Learning Road Traffic Detection in VR (AbnMLRTD-VR)’ prototype system was developed to assist the human analyst. The proposed system has two main integrative components: a data-driven ML model and a 3D real-time visualisation in a VR environment. An unsupervised ML model was built using real traffic data. The AbnMLRTD-VR system highlights the outliers in the road sections in actual road contexts of a road traffic network. This gives the human analyst a 3D real-time immersive visualisation in a VR environment to evaluate road conditions. The AbnMLRTD-VR system demonstrated that it could help minimise the need for human pre-labelling of the data. It enables the visualisation of the road traffic conditions more meaningfully and to understand the context of the road traffic conditions of road sections at any given time.

To improve users’ tendency towards shopping in Virtual Reality (VR), en-hancing the User Experience (UX) of the VR shopping environments is of primary importance. Product viewability, reachability, and personalisation are some of the primary UX factors in a shopping environment. This paper proposes and discusses three factors for a Personalised Adaptive Aisle (PAA) in a VR shopping environment to improve the shopping experience. They are 1) Shelf placement for viewability and reachability, 2) User view-point in VR, and 3) Personalised Product placement.

Gated Recurrent Neural Networks (RNNs) such as LSTM and GRU have been highly effective in handling sequential time series data in recent years. Although Gated RNNs have an inherent ability to learn complex temporal dynamics, there is potential for further enhancement by enabling these deep learning networks to directly use time information to recognise time-dependent patterns in data and identify important segments of time. Synonymous with time series data in real-world applications are missing values, which often reduce a model’s ability to perform predictive tasks. Historically, missing values have been handled by simple or complex imputation techniques as well as machine learning models, which manage the missing values in the prediction layers. However, these methods do not attempt to identify the significance of data segments and therefore are susceptible to poor imputation values or model degradation from high missing value rates. This paper develops Cyclic Gate enhanced recurrent neural networks with learnt waveform parameters to automatically identify important data segments within a time series and neglect unimportant segments. By using the proposed networks, the negative impact of missing data on model performance is mitigated through the addition of customised cyclic opening and closing gate operations. Cyclic Gate Recurrent Neural Networks are tested on several sequential time series datasets for classification performance. For long sequence datasets with high rates of missing values, Cyclic Gate enhanced RNN models achieve higher performance metrics than standard gated recurrent neural network models, conventional non-neural network machine learning algorithms and current state of the art RNN cell variants.

•A simple lipid extraction of chicken egg yolks was developed for LC-MS/MS analysis.

•937 lipid species from 20 major lipid subclasses were characterized in egg yolk.

•Statistical modeling was used to classify the types of conventional chicken eggs.

•Cage, barn, and free-range eggs can be differentiated based on lipid profile.

•Eggs from caged birds can be accurately predicted based on the lipidomic signature.

Free-range eggs are ethically desirable but as with all high-value commercial products, the establishment of provenance can be problematic. Here, we compared a simple one-step isopropanol method to a two-step methyl-tert-butyl ether method for extracting lipid species in chicken egg yolks before liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. The isopropanol method extracted 937 lipid species from 20 major lipid subclasses with high reproducibility (CV < 30 %). Machine learning techniques could differentiate conventional cage, barn, and free-range eggs using an external test dataset with an accuracy of 0.94, 0.82, and 0.82, respectively. Lipid species that differentiated cage eggs were predominantly phosphocholines and phosphoethanolamines whilst the free-range egg lipidomes were dominated by acylglycerides with up to three fatty acids. The lipid profiles were found to be characteristic of the cage, barns, and free-range eggs. The lipidomic analysis together with the statistical modeling approach thus provides an efficient tool for verifying the provenance of conventional chicken eggs.

Time series modelling has been successfully handled by Long Short-Term Memory (LSTM) models. Yet their performance can be severely inhibited by the occurrence of missing values prevalent in many real-life datasets. Many previous studies have been dedicated to imputation methods for generating a complete time series sequence, which have their limitations in terms of imputation bias and inaccuracies. In this paper, we propose a new LSTM model incorporating policy gradient (PG) based reinforcement learning called PG-LSTM, which can mitigate the effect of missing data and capture time-based input feature patterns more effectively to improve prediction performance. Inspired by numerous sequence models’ successes in improving the efficiency of processing language data by skipping irrelevant tokens, the PG-LSTM introduces dynamic skip connections between LSTM cell states for time series data for classification and regression tasks for the first time. Specifically, the proposed model comprises a modified LSTM cell architecture that can internally call a policy-based reinforcement learning agent to generate a skipping action, allowing the model to dynamically select the optimal subset of hidden and cell states from past states to capture periodic and non-periodic patterns within a time series sequence. Moreover, the PG-LSTM also designs a lightweight imputation layer using a simple missing value imputation strategy while incorporating missing indicators and skipping segments of unimportant data to reduce the limitations associated with imputed data for handling missing values. Our experimental results on regression and classification tasks on time series data with high rates of missing values demonstrate that the PG-LSTM improves performance against current gated recurrent neural networks (RNN) and conventional non-neural network algorithms. The PG-LSTM can enhance AUC by up to 18.5% in the classification task and RMSE by up to 19.3% in the regression task over gated RNN models, respectively. Our findings are also statistically analysed using statistical significance testing with post hoc analysis.

The continuously increasing renewable distributed generation (DG) penetration rate significantly reduces environmental pollution and power generation cost and satisfies society’s rapid growth in electricity demand. Nevertheless, high penetration of renewable DGs, such as wind power and photovoltaics (PV), might deteriorate the system’s efficiency and reliability due to its intermittent and stochastic natures. Introducing battery energy storage systems (BESSs) to the distribution system provides a practical method to compensate for the above deficiency since it can deliver and absorb power when needed. Hence, it is important to determine the optimal allocation of BESS to achieve maximum assistance in the grid. This study proposes an optimal BESS allocation methodology to improve reliability and economics in unbalanced distribution systems. The optimal BESS allocation problem is solved by simultaneously minimizing the cost of energy interruption, expected energy not supplied, power loss, line loading, voltage deviation, and BESS cost. The proposed technique is implemented and analyzed on a high renewable DG penetrated unbalanced IEEE-33 bus network using DIgSILENT PowerFactory software (version 2020 SP2A). An enhanced grey wolf optimization (EGWO) algorithm is developed to optimize BESS location and size according to the selected objective function. The simulation results show that the proposed optimal BESS optimization technique significantly improves the economics and reliability in unbalanced distribution systems and the EGWO outperforms the gray wolf optimization (GWO) and particle swarm optimization (PSO) algorithms.