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

Differentiable architecture search has gradually become the mainstreamresearch topic in the field of Neural Architecture Search (NAS) for its highefficiency compared with the early NAS methods. Recent differentiable NAS alsoaims at further improving the search performance and reducing the GPU-memoryconsumption. However, these methods are no longer naturally capable of tacklingthe non-differentiable objectives, e.g., energy, resource-constrainedefficiency, and other metrics, let alone the multi-objective search demands.Researches in the multi-objective NAS field target this but requires vastcomputational resources cause of the sole optimization of each candidatearchitecture. In light of this discrepancy, we propose the TND-NAS, which iswith the merits of the high efficiency in differentiable NAS framework and thecompatibility among non-differentiable metrics in Multi-objective NAS. Underthe differentiable NAS framework, with the continuous relaxation of the searchspace, TND-NAS has the architecture parameters been optimized in discretespace, while resorting to the progressive search space shrinking byarchitecture parameters. Our representative experiment takes two objectives(Parameters, Accuracy) as an example, we achieve a series of high-performancecompact architectures on CIFAR10 (1.09M/3.3%, 2.4M/2.95%, 9.57M/2.54%) andCIFAR100 (2.46M/18.3%, 5.46/16.73%, 12.88/15.20%) datasets. Favorably, comparedwith other multi-objective NAS methods, TND-NAS is less time-consuming (1.3GPU-days on NVIDIA 1080Ti, 1/6 of that in NSGA-Net), and can be convenientlyadapted to real-world NAS scenarios (resource-constrained,platform-specialized).

TND-NAS: Towards Non-differentiable Objectives in Progressive Differentiable NAS Framework