As the global population grows and more people migrate to urban areas, road traffic becomes an increasingly critical problem. Minimizing and mitigating the effects of disaster traffic are crucial for reducing emissions, economic losses, and saving lives by enabling faster emergency response times. Existing works aim to optimize traffic and rerouting in disaster-struck scenarios, but different cost functions are effective under different circumstances and information levels. No single method consistently produces an optimal rerouting cost function regardless of the road segment affected or the information provided. This study aims to develop a method to generate a cost function that consistently produces optimal solutions and outperforms existing functions, regardless of the road segment affected or the amount of information available. We simulated five different disaster scenarios in Dublin City Centre, with each scenario involving a disaster-struck road segment. We considered the use of Unmanned Aerial Vehicles to capture data on road conditions, dividing it into five information levels, with each level adding data about roads farther from the disaster. Furthermore, we used Grammar-Guided Genetic Programming (G3P) to evolve efficient cost functions, which were then tested in the SUMO simulator using Dijkstra's algorithm to reroute traffic. The performance was measured by the Average Travel Time of vehicles. Our approach successfully generated efficient cost functions for each scenario and information level. In all but one case, the G3P-generated functions outperformed existing methods. In some scenarios, the G3P-generated functions achieved better average arrival times than those in non-disaster conditions. We observed a maximum improvement of 44.23% in the Average Arrival Time compared to the no rerouting scenario and 14.99% compared to the non-disaster scenario. The findings suggest that our proposed G3P-based rerouting approach could be beneficial even in regular non-disaster situations.

Genetic algorithms (GA) allow us to optimise graphs according to multiple objectives while considering many different constraints. These population-based algorithms assess the fitness of a high number of genomes. In the case of optimising road networks, a high number of fitness assessments leads to high computational costs of traffic simulations. In this work, we explore the application of learning graph configuration spaces to make efficient use of these traffic simulations by using learning model predictions for the majority of fitness assessments. In a controlled experiment, we compare the quality of GA optimisations with and without learning model predictions on the same simulation budget. Our results indicate that although we lose accuracy in the fitness assessments with predictions, the GA reliably finds road networks with better traffic flow and lower overall road length while using the same number of traffic simulations. We show that learning models can support GAs to make efficient use of the simulation budget and thus improve the optimisation. Future work is necessary to confirm these results for larger road networks.

As one of the most detrimental code smells, code clones significantly increase software maintenance costs and heighten vulnerability risks, making their detection a critical challenge in software engineering. Abstract Syntax Trees (ASTs) dominate deep learning-based code clone detection due to their precise syntactic structure representation, but they inherently lack semantic depth. Recent studies address this by enriching AST-based representations with semantic graphs, such as Control Flow Graphs (CFGs) and Data Flow Graphs (DFGs). However, the effectiveness of various enriched AST-based representations and their compatibility with different graph-based machine learning techniques remains an open question, warranting further investigation to unlock their full potential in addressing the complexities of code clone detection. In this paper, we present a comprehensive empirical study to rigorously evaluate the effectiveness of AST-based hybrid graph representations in Graph Neural Network (GNN)-based code clone detection. We systematically compare various hybrid representations ((CFG, DFG, Flow-Augmented ASTs (FA-AST)) across multiple GNN architectures. Our experiments reveal that hybrid representations impact GNNs differently: while AST+CFG+DFG consistently enhances accuracy for convolution- and attention-based models (Graph Convolutional Networks (GCN), Graph Attention Networks (GAT)), FA-AST frequently introduces structural complexity that harms performance. Notably, GMN outperforms others even with standard AST representations, highlighting its superior cross-code similarity detection and reducing the need for enriched structures.

The fairness of Artificial Intelligence (AI) has gained tremendous attention within the criminal justice system in recent years, mainly when predicting the risk of recidivism. The primary reason is attributed to evidence of bias towards demographic groups when deploying these AI systems. Many proposed fairness-improving techniques applied at each of the three phases of the fairness pipelines, pre-processing, in-processing and post-processing phases, are often ineffective in mitigating the bias and attaining high predictive accuracy. This paper proposes a novel approach by integrating existing fairness-improving techniques: Reweighing, Adversarial Learning, Disparate Impact Remover, Exponential Gradient Reduction, Reject Option-based Classification, and Equalized Odds optimization across the three fairness pipelines simultaneously. We evaluate the effect of combining these fairness-improving techniques on enhancing fairness and attaining accuracy. In addition, this study uses multi- and bi-objective optimization techniques to provide and to make well-informed decisions when predicting the risk of recidivism. Our analysis found that one of the most effective combinations (i.e., disparate impact remover, adversarial learning, and equalized odds optimization) demonstrates a substantial enhancement and balances achievement in fairness through various metrics without a notable compromise in accuracy.

Intelligent computational methods for economicsEconomics explores the behavior of people, companies, governments, and various decision-makers to explain how their decisions produce value and satisfy (or fail to satisfy) human needs and desires.Driven by advances in artificial intelligence (AI) and reinforced by the acumen of generated and collected open data, there is now a significant and growing field that utilizes the concept of a synthetic homo economicus, the mythical per-

Artificial Intelligence (AI) is applied in almost every public sector because of its positive impacts. However, AI’s ethical aspects and trustworthiness constitute a significant uproar and concern among different AI stakeholders due to AI’s adverse effect on users when the AI system lacks cautionary measures. AI is used in the criminal justice system for predicting recidivism risk. However, AI’s negative impact translates into bias and high incarceration towards a group of defendants in a population assessed for recidivism risk. This paper focuses on fairness as a requirement of a trustworthy AI framework previously proposed to ascertain the appropriate application of AI systems in predicting recidivism. This paper aims to raise awareness about the fairness of AI models and stimulate further research and deployment of efficient and effective exploitation of fair and trustworthy AI models in the criminal justice system when predicting recidivism. Fairness has been a significant concern for criminal justice system stakeholders and has received considerable attention with more theoretical and practical studies than other trustworthy AI requirements. Hence, this paper reviews state-of-the-art fairness, outlines valuable findings, and proposes future directions to achieve fair AI systems for predicting recidivism risk. In addition, this paper ensures mapping existing technical works in the literature to the fairness pipeline corresponding to the criminal justice system’s AI development phases.

New blockchain platforms are launching at a high cadence, each fighting for attention, adoption, and infrastructure resources. Several studies have measured the peer-to-peer (P2P) network decentralisation of Bitcoin and Ethereum (i.e., two of the largest used platforms). However, with the increasing demand for blockchain infrastructure, it is important to study node decentralisation across multiple blockchain networks, especially those containing a small number of nodes. In this paper, we propose NodeMaps, a data processing framework to capture, analyse, and visualise data from several popular P2P blockchain platforms, such as Cosmos, Stellar, Bitcoin, and Lightning Network. We compare and contrast the geographic distribution, the hosting provider diversity, and the software client variance in each of these platforms. Through our comparative analysis of node data, we found that Bitcoin and its Lightning Network Layer 2 protocol are widely decentralised P2P blockchain platforms, with the largest geographical reach and a high proportion of nodes operating on The Onion Router (TOR) privacy-focused network. Cosmos and Stellar blockchains have reduced node participation, with nodes predominantly operating in large cloud providers or well-known data centres.

Grammar-Guided Genetic Programming (G3P) is widely recognised as one of the most successful approaches for program synthesis, i.e., the task of automatically discovering an executable piece of code given user intent. G3P has been shown capable of successfully evolving programs in arbitrary languages that solve several program synthesis problems based only on a set of input/output examples. Despite its success, the restriction on the evolutionary system to only leverage input/output error rate during its assessment of the programs it derives limits its scalabil-ity to larger and more complex program synthesis problems. With the growing number and size of open software repositories and generative artificial intelligence approaches, there is a sizeable and growing number of approaches for retrieving/generating source code (potentially several partial snippets) based on textual problem descriptions. Therefore, it is now, more than ever, time to introduce G3P to other means of user intent (particularly textual problem descriptions). In this paper, we would like to assess the potential for G3P to evolve programs based on their similarity to particular target codes of interest (obtained using some code retrieval/generative approach). Through our experimental evaluation on a well-known program synthesis benchmark, we have shown that G3P successfully manages to evolve some of the desired programs with all four considered similarity measures. However, in its default configuration, G3P is not as successful with similarity measures as it is with the classical input/output error rate when solving program synthesis problems. Therefore, we propose a novel multi-objective G3P approach that combines the similarity to the target program and the traditional input/output error rate. Our experiments show that compared to the error-based G3P, the multi-objective G3P approach could improve the success rate of specific problems and has great potential to improve on the traditional G3P system.

Grammar-Guided Genetic Programming is already outperforming humans at creating efficient transmission schedulers for large heterogeneous communications networks. We have previously proposed a multi-level grammar approach which achieved significantly better results than the canonical Grammar-Guided Genetic Programming approach. Initially, a restricted `small' grammar is utilised in order to discover suitable structures. A full grammar is then adopted after this initial phase. Hence, evolution can focus on maximising performance, by fine-tuning the well-structured models. In this work, we propose to use a hierarchical approach by employing multiple small grammars instead of a unique small grammar at the lower level, in conjunction with the full grammar at the upper level. To use multiple small grammars while maintaining the same computational budget, we have to use either (i) reduce the number of generations, or (ii) reduce the size of the population for the evolution with each of the small grammars. In this work, we confirm that the hierarchical grammar approach using the division of number of generations strategy achieves significantly better results than the multi-level approach, but requires defining an ideal number of small grammars to achieve the best performance. We also show that the hierarchical grammar approach using the division of population size strategy achieves significantly better results than the multi-level approach. However the division of population size strategy is less sensitive to the number of small grammars.

The number of drivers using on-board systems to navigate through urban areas is increasing. Drivers get real time information regarding traffic conditions and change their routes accordingly. Adapting a route clearly enables drivers to avoid closed roads or circumvent major hotspots. However, given the non-linearity of the traffic dynamics in urban environments, choosing a route based only on current traffic load or current average vehicle speed is not a guaranty of a lower overall travel time. In this work, we design an evolutionary system to search for better surrogate travel cost that drivers could optimise in their rerouting to achieve better overall travel times. Our system uses the Grammar-Guided Genetic Programming algorithm to evolve surrogate travel cost expressions and evaluate their performances on a micro traffic simulator. Our system is able to evolve different expressions that meet characteristics of specific urban environments instead of a one size fits all expression. We have seen in our experimental study on a traffic scenario representing Dublin city centre that our system is able to evolve surrogate travel cost expressions with ~34% and ~10% improvements in average travel time over the no rerouting and the average travel speed based rerouting algorithms.

Wireless communications networks are operating at breaking point during an era of relentless traffic growth. Network operators must utilize scarce and expensive wireless spectrum efficiently in order to satisfy demand. Spectrum on the links between cells and user equipments ('users': smartphones, tablets, etc.) frequently becomes congested. Capacity can be increased by transmitting data packets via multiple links. Packets can be routed through multiple Long Term Evolution (LTE) links in existing fourth generation (4G) networks. In future 5G deployments, users will be equipped to receive packets over LTE, WiFi, and millimetre wave links simultaneously.

Testing is crucial to ensure the quality of software systems-but testing is an expensive process, so test managers try to minimise the set of tests to run to save computing resources and speed up the testing process and analysis. One problem is that there are different perspectives on what is a good test and it is usually not possible to compare these dimensions. This is a perfect example of a multi-objective optimisation problem, which is hard-especially given the scale of the search space here. In this paper, we propose a novel hybrid algorithm to address this problem. Our method is composed of three steps: a greedy algorithm to find quickly some good solutions, a genetic algorithm to increase the search space covered and a local search algorithm to refine the solutions. We demonstrate through a large scale empirical evaluation that our method is more reliable (better whatever the time budget) and more robust (better whatever the number of dimensions considered)-in the scenario with 4 objectives and a default execution time, we are 178% better in hypervolume on average than the state-of-the-art algorithms.

Machine Reassignment is a challenging problem for constraint programming (CP) and mixed integer linear programming (MILP) approaches, especially given the size of data centres. The multi-objective version of the Machine Reassignment Problem is even more challenging and it seems unlikely for CP or MILP to obtain good results in this context. As a result, the first approaches to address this problem have been based on other optimisation methods, including metaheuristics. In this paper we study under which conditions a mixed integer optimisation solver, such as IBM ILOG CPLEX, can be used for the Multi-objective Machine Reassignment Problem. We show that it is useful only for small or medium scale data centres and with some relaxations, such as an optimality tolerance gap and a limited number of directions explored in the search space. Building on this study, we also investigate a hybrid approach, feeding a metaheuristic with the results of CPLEX, and we show that the gains are important in terms of quality of the set of Pareto solutions (+126.9% against the metaheuristic alone and +17.8% against CPLEX alone) and number of solutions (8.9 times more than CPLEX), while the processing time increases only by 6% in comparison to CPLEX for execution times larger than 100 seconds.

More and more drivers use on-board units to help them navigate in the increasing urbanised environment they live and work in. These system (e.g., routing applications on smart phones) are now very often on-line, and use information from the traffic situation (e.g., accidents, congestion) to get the best route. We can now envisage a world where all trips are assigned and updated by such an on-line system, making the best routing decisions based on traffic conditions. The problem is that current systems consider only 'local' elements (e.g., driver preference and current traffic condition) and do not make routing decisions from a global perspective. This can lead to a lot of similar routing assignments that could lead to further traffic congestion. The objective of the next generation on-line navigation systems is then to come up with a 'smart', real-time route assignment, which balances the load between the different road segments and offers the best quality to the drivers. However, every routing decision made has an impact on the traffic conditions (one more vehicle on the road segments selected) and computing the load induced by the trips is a computationally heavy problem. This paper addresses this question of real-time on-line traffic assignment, and shows that under certain conditions it is possible to have (i) an accurate estimation of the load and travel time on every road segment and (ii) an optimised traffic assignment that adapts to divergence and evolutions (e.g., accidents) of the system.