This paper proposes Shoggoth, an efficient edge-cloud collaborative architecture, for boosting inference performance on real-time video of changing scenes. Shoggoth uses online knowledge distillation to improve the accuracy of models suffering from data drift and offloads the labeling process to the cloud, alleviating constrained resources of edge devices. At the edge, we design adaptive training using small batches to adapt models under limited computing power, and adaptive sampling of training frames for robustness and reducing bandwidth. The evaluations on the realistic dataset show 15%-20% model accuracy improvement compared to the edge-only strategy and fewer network costs than the cloud-only strategy.

His work has appeared in The Lovecraft eZine, Samsara: The Magazine of Suffering, Tigershark eZine, Turn To Ash horror zine, The Atlantean Supplement, The Eldritch Literary Review, The Chamber, and Horizontum (Mexico City). John's first novel in the Dark Union series, NIGHT OF THE KWATEE is now available (on Amazon), published by Night Horse Publishing House. His horror short, "The Thing Beneath the Tree," also appears in the PROTECTORS OF THE VEIL anthology from the Lovecraft Lunatic Society (on Amazon). Follow John's latest publication news on Twitter @HPL_JDeLaughter or Facebook @HPLJDeLaughter.