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/

Zendy Plus

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

Zendy Tools

Zendy Open

DNN learning jobs are common in today's clusters due to the advances in AIdriven services such as machine translation and image recognition. The mostcritical phase of these jobs for model performance and learning cost is thetuning of hyperparameters. Existing approaches make use of techniques such asearly stopping criteria to reduce the tuning impact on learning cost. However,these strategies do not consider the impact that certain hyperparameters andsystems parameters have on training time. This paper presents PipeTune, aframework for DNN learning jobs that addresses the trade-offs between these twotypes of parameters. PipeTune takes advantage of the high parallelism andrecurring characteristics of such jobs to minimize the learning cost via apipelined simultaneous tuning of both hyper and system parameters. Ourexperimental evaluation using three different types of workloads indicates thatPipeTune achieves up to 22.6% reduction and 1.7x speed up on tuning andtraining time, respectively. PipeTune not only improves performance but alsolowers energy consumption up to 29%.

PipeTune: Pipeline Parallelism of Hyper and System Parameters Tuning for Deep Learning Clusters