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

The explosive growth of various types of big data and advances in AItechnologies have catalyzed a new type of workloads called multi-modal DNNs.Multi-modal DNNs are capable of interpreting and reasoning about informationfrom multiple modalities, making them more applicable to real-world AIscenarios. In recent research, multi-modal DNNs have outperformed the bestuni-modal DNN in a wide range of distributed computing applications fromtraditional multimedia systems to emerging autonomous edge systems. However,despite their importance and superiority, very limited research attention hasbeen devoted to understand the characteristics of multi-modal DNNs and theirimplications on current computing software/hardware platforms. Existingbenchmarks either target uni-modal DNNs or only focus on the algorithmcharacteristics of multi-modal DNNs. There lacks representative benchmarksuites that provide comprehensive system and architecture level analysis ofmulti-modal networks.  To advance the understanding of these multi-modal DNN workloads andfacilitate related research, we present MMBench, an open-source, end-to-endbenchmark suite consisting of a set of real-world multi-modal DNN workloadswith relevant performance metrics for evaluation. We then use MMBench toconduct an in-depth analysis on the characteristics of multi-modal DNNs. Wedemonstrate their unique characteristics of clear multi-stage execution,frequent synchronization and high heterogeneity, which distinguish them fromconventional uni-modal DNNs. Finally, we conduct a case study and extend ourbenchmark to edge devices. We hope that our work can provide insights forfuture software/hardware design and optimization to underpin multi-modal DNNson both cloud and edge computing platforms.

MMBench: Benchmarking End-to-End Multi-modal DNNs and Understanding Their Hardware-Software Implications