English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Continual learning (CL) trains NN models incrementally from a continuousstream of tasks. To remember previously learned knowledge, prior studies storeold samples over a memory hierarchy and replay them when new tasks arrive. Edgedevices that adopt CL to preserve data privacy are typically energy-sensitiveand thus require high model accuracy while not compromising energy efficiency,i.e., cost-effectiveness. Our work is the first to explore the design space ofhierarchical memory replay-based CL to gain insights into achievingcost-effectiveness on edge devices. We present Miro, a novel system runtimethat carefully integrates our insights into the CL framework by enabling it todynamically configure the CL system based on resource states for the bestcost-effectiveness. To reach this goal, Miro also performs online profiling onparameters with clear accuracy-energy trade-offs and adapts to optimal valueswith low overhead. Extensive evaluations show that Miro significantlyoutperforms baseline systems we build for comparison, consistently achievinghigher cost-effectiveness.

Cost-effective On-device Continual Learning over Memory Hierarchy with Miro