Rajkumar Buyya

LLM scaling bottlenecks demand a shift towards cloud-native architectures and distributed systems, unlocking potential gains from serverless inference and quantum computing.

Like the periodic table for chemistry, a new "periodic framework" promises to bring order and predictability to the chaotic world of distributed computing.

Quantum reinforcement learning gets a distributed boost, achieving 10% better performance in multi-agent environments by distributing the learning load across multiple quantum agents.

Federated reinforcement learning can now handle heterogeneous, adversarial IoT environments with near-zero deadline violations, thanks to a novel decentralized framework that transfers knowledge across silos.

Overcome the scalability bottleneck of GNN-based root cause analysis in edge computing by cascading subnetworks over clustered service graphs, achieving near-constant inference latency without sacrificing accuracy.

Optimizing UAV routes and edge computing together slashes wildfire monitoring response times by up to 84% while shrinking drone fleets by over 40%.

Circuit cutting introduces substantial end-to-end overheads in quantum neural network training, with reconstruction dominating per-query time, but surprisingly, test accuracy and robustness can be preserved or even improved.