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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Dataflow/mapping decides the compute and energy efficiency of DNNaccelerators. Many mappers have been proposed to tackle the intra-layermap-space. However, mappers for inter-layer map-space (aka layer-fusionmap-space), have been rarely discussed. In this work, we propose a mapper,DNNFuser, specifically focusing on this layer-fusion map-space. While existingSOTA DNN mapping explorations rely on search-based mappers, this is the firstwork, to the best of our knowledge, to propose a one-shot inference-basedmapper. We leverage Transformer as our DNN architecture to learn layer-fusionoptimization as a sequence modeling problem. Further, the trained DNNFuser cangeneralize its knowledge and infer new solutions for unseen conditions. Withinone inference pass, DNNFuser can infer solutions with compatible performance tothe ones found by a highly optimized search-based mapper while being 66x-127xfaster.

DNNFuser: Generative Pre-Trained Transformer as a Generalized Mapper for Layer Fusion in DNN Accelerators