This paper presents an RL-based auto-scaler for container workloads across edge, fog, and cloud tiers. It dynamically adjusts instances based on latency, bandwidth, and CPU pressure, ensuring optimal resource utilization and SLA compliance in decentralized container infrastructures.

Ayaan Raghav Choudhury, Lucas Henry Whitaker, Chen Rong Xiang, Ava Grace Holloway, Martina Eloisa Gomez

This work proposes a semantic AI engine to enforce governance policies across geographically distributed cloud resources. It supports real-time detection of violations and remediates actions automatically through intelligent mapping of compliance rules to dynamic infrastructure configurations.

Charlotte Evelyn Bishop, Pratik Mohan Srinivasan, Haoran Qiu Wei, Thomas Finley Cross, Zoe Amalia Richardson, Diego Felipe Morales

This paper introduces a secure workload auditing mechanism using homomorphic encryption for AI tasks in federated data lakes. It allows analytics without exposing raw data, maintaining confidentiality while enabling performance benchmarking and regulatory compliance across cloud-based AI pipelines.

Liang Hao Sheng, Amelia Hope Thornton, Tanmay Ishaan Menon, Hugo Bernard Davies, Lucia Teresa Castillo

This study presents a GNN-based fault predictor for proactive issue detection in cloud orchestration. It captures topological dependencies among services and predicts impending node failures, supporting self-healing mechanisms in multi-tenant deployments without impacting operational latency.

Sofia Isabelle Carter, Yu Xin Bo, Aarush Devendra Rathi, Marcus Elijah Walker, Isabella Josephine Owen

This paper proposes a federated learning-based framework to optimize resource scheduling for smart cloud gateways. It minimizes bandwidth consumption by predicting edge workload profiles and redistributing tasks across gateways using decentralized cognitive scheduling agents.

Huang Wenjie Xu, Isabella May Barrett, Nitin Sharath Rao, Samuel Leo Donovan, Francesca Isabel Russo