Peter Bloodsworth

Jingni Huang, Peter Bloodsworth

Short-term human pose prediction plays a crucial role in interactive systems, assistive robots, and emotion-aware human-computer interaction[1-3]. While current trajectory prediction models primarily rely on geometric motion cues, they often overlook the underlying emotional signals influencing human motion dynamics[4-5]. This paper investigates whether facial expression-derived emotion embeddings can provide auxiliary conditional signals for short-term pose prediction. To further evaluate multimodal conditionation in a recursive prediction setting, we propose a lightweight autoregressive predictive world model that performs 15-step rolling pose prediction. This framework combines pose keypoints with emotion embeddings through a learnable gating mechanism and performs autoregressive unfolding prediction using a recurrent sequence model based on a two-layer LSTM architecture. Experiments were conducted on two small-scale pose-emotion video datasets: controlled motion sequences with minimal facial expression changes and, natural emotion-driven motion sequences with considerable facial expression changes. The results show that simple multimodal fusion does not consistently improve prediction accuracy, while normalized gating fusion significantly enhances the performance of emotion-driven motion sequences. Furthermore, counterfactual perturbation experiments demonstrate that the predicted trajectory exhibits measurable sensitivity to changes in multimodal input, suggesting that facial expression embeddings act as auxiliary conditional signals rather than redundant features. In summary, these results indicate that incorporating facial expression-derived emotion embeddings into emotion-conditional short-term pose prediction based on a lightweight predictive world model architecture is a feasible approach.

Nauman Khalid, Ghalib Ahmed Tahir, Peter Bloodsworth

Multi-agent systems are often limited in terms of persistenceand scalability. This issue is more prevalent for applications inwhich agent states changes frequently. This makes the existingmethods less usable as they increase the agent's complexityand are less scalable. This research study has presented anovel in-memory agent persistence framework. Two prototypeshave been implemented, one using the proposed solution andthe other using an established agent persistency environment.Experimental results confirm that the proposed framework ismore scalable than existing approaches whilst providing asimilar level of persistency. These findings will help futurereal-time multiagent systems to become scalable and persistentin a dynamic cloud environment.

Syeda ZarAfshan Gohera, Peter Bloodsworth, Raihan Ur Rasool et al.

Growing interest in Cloud Computing places a heavy workload on cloud providers which is becoming increasingly difficult for them to manage with their primary datacenter infrastructures. Resource limitations can make providers vulnerable to significant reputational damage and it often forces customers to select services from the larger, more established companies, sometimes at a higher price. Funding limitations, however, commonly prevent emerging and even established providers from making continual investment in hardware speculatively assuming a certain level of growth in demand. As an alternative, they may strive to use the current inter-cloud resource sharing platforms which mainly rely on monetary payments and thus putting pressure on already stretched cash flows. To address such issues, we have designed and implemented a new multi-agent based Cloud Resource Bartering System (CRBS) that fosters the management and bartering of pooled resources without requiring costly financial transactions between providers. Agents in CRBS not only strengthen the trading relationship among providers but also enable them to handle surges in demand with their primary setup. Unlike existing systems, CRBS assigns resources by considering resource urgency which comparatively improves customers satisfaction and the resource utilization rate by more than 50%.The evaluation results provide evidence that our system assists providers to timely acquire the additional resources and to maintain sustainable service delivery. We conclude that the existence of such a system is economically beneficial for cloud providers and enables them to adapt to fluctuating workloads.

R. McClatchey, A. Branson, A. Anjum et al.

With the increasingly digital nature of biomedical data and as the complexity of analyses in medical research increases, the need for accurate information capture, traceability and accessibility has become crucial to medical researchers in the pursuance of their research goals. Grid- or Cloud-based technologies, often based on so-called Service Oriented Architectures (SOA), are increasingly being seen as viable solutions for managing distributed data and algorithms in the bio-medical domain. For neuroscientific analyses, especially those centred on complex image analysis, traceability of processes and datasets is essential but up to now this has not been captured in a manner that facilitates collaborative study. Over the past decade, we have been working with mammographers, paediatricians and neuroscientists in three generations of projects to provide the data management and provenance services now required for 21st century medical research. This paper outlines the finding of a requirements study and a resulting system architecture for the production of services to support neuroscientific studies of biomarkers for Alzheimers Disease. The paper proposes a software infrastructure and services that provide the foundation for such support. It introduces the use of the CRISTAL software to provide provenance management as one of a number of services delivered on a SOA, deployed to manage neuroimaging projects that have been studying biomarkers for Alzheimers disease.

Ashiq Anjum, Peter Bloodsworth, Andrew Branson et al.

We outline the approach being developed in the neuGRID project to use provenance management techniques for the purposes of capturing and preserving the provenance data that emerges in the specification and execution of workflows in biomedical analyses. In the neuGRID project a provenance service has been designed and implemented that is intended to capture, store, retrieve and reconstruct the workflow information needed to facilitate users in conducting user analyses. We describe the architecture of the neuGRID provenance service and discuss how the CRISTAL system from CERN is being adapted to address the requirements of the project and then consider how a generalised approach for provenance management could emerge for more generic application to the (Health)Grid community.

Ashiq Anjum, Peter Bloodsworth, Irfan Habib et al.

By abstracting Grid middleware specific considerations from clinical research applications, re-usable services should be developed that will provide generic functionality aimed specifically at medical applications. In the scope of the neuGRID project, generic services are being designed and developed which will be applied to satisfy the requirements of neuroscientists. These services will bring together sources of data and computing elements into a single view as far as applications are concerned, making it possible to cope with centralised, distributed or hybrid data and provide native support for common medical file formats. Services will include querying, provenance, portal, anonymization and pipeline services together with a 'glueing' service for connection to Grid services. Thus lower-level services will hide the peculiarities of any specific Grid technology from upper layers, provide application independence and will enable the selection of 'fit-for-purpose' infrastructures. This paper outlines the design strategy being followed in neuGRID using the glueing and pipeline services as examples.