African Institute for Mathematical Sciences

The study addresses the energy challenges faced by Lomé, where electricity demand often exceeds supply, leading to frequent load-shedding and exacerbating inequalities in access among sub-zones. It focuses on optimizing the fair distribution of electricity by combining the cake-cutting algorithm with artificial intelligence (AI) models. An analysis of existing heuristic algorithms (Random-Selector Algorithm RSA, Grouper Algorithm GA, Consumption-Sorter Algorithm CSA1, and Cost-Sorter Algorithm CSA2) revealed their limitations in ensuring equity. To address these shortcomings, a hybrid approach was developed, integrating dynamic parameters and convolutional LSTM networks coupled with wavelet transforms. Tests conducted over periods ranging from 1 to 7 days demonstrated significant performance improv ements. For instance, over 7 days, the hybrid approach (α = 0.8, β = 0.5) reduced the maximum connection gap to 20 hours, compared to 105 hours for CSA1 and 91 hours for CSA2. The proposed solution offers an efficient and equitable method for electricity d istribution, reducing inequalities while supporting the sustainable management of urban energy resources.

The world’s natural gas and crude oil resources are diminishing to the point that gas and petroleum prices are lowering the common man’s standard of living worldwide. Likewise, there is a need for the world to move towards a carbonless energy future to curb the influence of climate change, with increasingly intense climate phenomena observed every year. Subsequently, the search continues for increasingly practical renewable energy sources and effective energy storage methods. Hydrogen, as a clean and abundant energy source, has the highest energy density in terms of mass when excluding nuclear fuels. In terms of the United Nations' Sustainable Development Goals, hydrogen will play a crucial role, specifically in achieving Goal 7, Affordable and Clean Energy, and Goal 13, Climate Action. This study seeks to lay out the current knowledge base of hydrogen energy, focusing on solid-state hydrogen storage. The focus on solid-state hydrogen storage is due to the need for long-term safe storage, which is one of the challenges experienced by hydrogen energy and the hydrogen economy. Moreover, solid-state hydrogen storage offers a potential solution to this challenge. The analysis outlined the current state of hydrogen energy as well as the need for further investigation into metal-hydride-based hydrogen storage while considering the benefits that simulation and machine learning have shown in the field of research. In so doing, it justifies further experimental and simulation-based research into solid-state hydrogen storage using hydride-forming metals. This paper shows that there is not enough published work on large-scale solid-phase hydrogen storage units, as well as the lack of operational machine learning modelling of hydrogen systems. • Deep analysis of metal hydride mechanism • Analysis of optimisation literature. • Analysis of simulation studies • Outlining knowledge gap

This paper empirically tests the Environmental Kuznets Curve (EKC) hypothesis for sub-Saharan Africa, using CO₂ emission as a proxy for environmental quality. The analysis employs panel pooled data for 39 countries from 1990 to 2019. We find evidence of an inverted U-shaped relationship between income and emissions, as GDP per capita has a positive effect and its square has a negative effect on CO₂ emissions, thereby supporting the EKC hypothesis. Our results reveal a dual role for renewable energy. When CO₂ emissions are the dependent variable, renewable energy consumption significantly reduces emissions, while non-renewable energy increases them. Additionally, population size and literacy rate are found to positively affect CO₂ emissions. In contrast, when economic growth is the dependent variable, renewable energy has a positive and significant influence. These findings indicate that expanding renewable energy production in sub-Saharan Africa offers a dual advantage, enhancing environmental quality while also contributing to stronger economic performance. By demonstrating that renewable energy reduces CO₂ emissions and supports growth, the results provide important guidance for policymakers seeking to advance sustainable development pathways that are compatible with long‑term environmental conservation across the region.

Abstract In this paper, we introduce a nonzero component graph of a finite-dimensional near-vector space constructed from a finite field. We investigate the relationship between the existence of a basis of a near-vector space over a field with that of the vector space over the same field, with the same dimension. We introduce the notion of a coordinate vector and prove an equivalent formulation for the definition of the nonzero component graph. We show that despite the vast differences in the structure of a near-vector space compared with a vector space, especially in the nonregular case, the graph fails the recognition problem in that it cannot distinguish between the near-vector space and its associated vector space. Thus, we show that the nonzero component graph of a near-vector space is identical to the one for a vector space over the same field, of the same dimension. Leveraging this result, we completely characterise and describe graph properties for the nonzero component graph on both near-vector spaces and vector spaces by partitioning the graph into components using the regularity graph of any nonregular near-vector space (with the same dimension) with its nonzero component graph. In closing we provide a reconstruction algorithm for constructing a near-vector space for a given nonzero component graph, satisfying certain conditions.

ABSTRACTIn an increasingly fragmented natural world, understanding how different ecological phenomena vary with patch size has many motivations. Examples include the assembly of biodiversity, ecosystem service provision and the suitability of fragments for habitat specialist species. A common approach to such questions divides fragments into small and large size classes for separate analysis. However, lack of an objective definition and means to differentiate ‘small’ from ‘large’ patches limits our ability to compare findings across studies, arguably impeding progress toward any unified views. Because larger and smaller fragments tend, on average, to respectively over‐represent narrow‐ and wide‐range species, an ‘area for unbiased species representation’ (AUSR) can be defined at some intermediate fragment size predicted to contain species at incidence frequencies approximating that of the overall landscape. A central tendency for AUSR has previously been estimated for patchy habitats (islands, habitat islands and fragments), providing a benchmark to compare this threshold of small fragment size between studies. However, if AUSR can be readily determined within individual study systems, it would also provide an objective threshold to separate small and large fragments under the AUSR definition. Here we assess this potential for 138 published datasets from various fragmented landscapes using an index comparing species incidence frequencies in each fragment with that of the overall landscape. Regressing this index on fragment area yielded an estimate for AUSR in over 90% of cases, suggesting broad applicability as an objective way to separate fragments into two size classes. Regression slopes provide further information on the relative representation of narrow‐ vs. wide‐range species, with ~80% being numerically consistent with the overall negative trend. Requiring only the same data as the island species‐area relationship, AUSR can provide useful insights on the relative importance of narrow‐ vs. wide‐ranging species for studies of patch‐size dependence in ecological phenomena.

Background/Objectives: Emerging evidence suggests that multiple micronutrient supplements (MMS) provide additional benefits for maternal and neonatal health compared with iron and folic acid (IFA) supplements. To achieve effective coverage, acceptability, and adherence—and to inform a nationwide rollout of MMS—it is essential to understand the context-specific factors that shape implementation. This study evaluated the pilot implementation of MMS in Rwanda to identify key enablers, areas for improvement, and challenges related to antenatal care (ANC) attendance and MMS use. Methods: Data were collected through a survey of 3257 women who attended ANC services, seven focus group discussions with 35 ANC attendees, and key informant interviews with 20 ANC nurses and 21 community health workers. Results: Pregnant women reported high ANC attendance (74%) and MMS consumption (79%), largely driven by strong motivation and awareness of MMS benefits. Strategies to remember daily intake and to manage side effects supported adherence, as did reminders, motivation, and information from family members and healthcare providers. Limited patient-centered counselling, financial constraints, barriers to accessing ANC services, and product stock-outs were key areas for strengthening service delivery in Rwanda. Conclusions: Sustaining high ANC attendance and MMS adherence as the program transitions from the pilot phase to national scale-up is essential. Improving counseling quality and strengthening supply chains may reinforce ANC services and support sustained MMS adherence, with benefits for maternal and child health.

Abstract Human language enables the exchange of complex information and precise instructions for collaborative planning and action. It rapidly evolves through social learning, generating diverse cultural communication signals used not only with other humans, but also with domesticated animals bred or trained to respond. More rarely, humans communicate with untrained, wild animals to coordinate joint actions, yet little is known about how or why these human‐to‐wildlife signals diversify. Human‐wildlife cooperation allows us to investigate whether human signals directed at untrained, wild animals exhibit regional variation, akin to dialects in human language. We investigated regional variation in human signals used to cooperate with greater honeyguides ( Indicator indicator ): wild birds that guide people to bees' nests in exchange for access to beeswax. Across 13 villages in northern Mozambique, we tested whether human honey‐hunting calls varied with spatial distance, as expected if regional dialects had emerged, or with measures of the physical environment affecting sound propagation, as expected if calls were shaped by habitat acoustics. Our analyses showed that trills, grunts, whoops, and whistles used while cooperating with honeyguides (i) consistently varied with spatial distance between villages, (ii) varied irrespective of the local habitat, and (iii) appeared to be adopted by immigrant honey‐hunters to match local calls. These findings suggest that regional variation in human‐to‐wildlife signals is shaped primarily by human social factors, forming a landscape of interspecific signal diversity similar to human language dialects. Honeyguides cooperate effectively with honey‐hunters throughout this landscape, suggesting that they accommodate (and likely reinforce) cultural differences by learning the local interspecies dialect. Read the free Plain Language Summary for this article on the Journal blog.

Thermal reaction norms, or thermal performance curves (TPCs), describe how ectothermic organisms respond to temperature variation. Here, we investigate how stochastic environmental noise, by modulating competitive interactions among individuals with differing TPC shapes, can drive long-term evolutionary changes in thermal performance. We develop a Ricker competition model within the framework of adaptive dynamics to examine how environmental variability, corresponding to the joint effect of both periodic and stochastic temperature fluctuations, influences the evolution of competition TPCs. Growth, carrying capacity and competitive interactions are all temperature-dependent, with competition coefficients defined by the ratio of individuals’ thermal performance curve at the prevailing temperature. Competition TPCs are modelled using a beta probability density function. Previous results from periodically (i.e., deterministically) fluctuating environments established that the thermal optimum converges to the mean environmental temperature, while performance breadth collapses under constant conditions. In contrast, we show that under stochastic environments, increasing thermal noise broadens competition TPCs and shifts the thermal optimum leftward, away from the environmental mean. This results in right-skewed competition TPCs and suggests an evolutionary bias towards cooler temperature optima for competition under moderate to high environmental noise. Alongside the broadening of competition TPCs to buffer against thermal fluctuations, the shift towards cooler optima reflects a more conservative thermal strategy for competition performance in the face of environmental uncertainty.Supplementary InformationThe online version contains supplementary material available at 10.1007/s11538-026-01614-6.

Efficient spectrum allocation is vital for 5G/6G networks, yet traditional deep reinforcement learning (DRL) methods suffer from high sample complexity and unsafe exploration that can disrupt network stability. To address these challenges, we propose a meta-learning framework that learns a robust initial policy capable of rapid and safe adaptation to changing wireless conditions. We implement three meta-learning architectures using model-agnostic techniques—model-agnostic meta-learning (MAML), recurrent neural network (RNN), and RNN with a self-attention mechanism—and compare them against a DRL baseline and classical heuristic approaches in a dynamic integrated access/ backhaul (IAB) environment. The attention-based agent achieves a peak throughput of ≈ 49 Mbps, reducing SINR and latency violations by over 60% relative to PPO, and attains 97% of the fairness level of the exhaustive-search upper bound. These results demonstrate that meta-learning enables data-efficient, reliable, and scalable spectrum management for next-generation wireless systems.