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

Neural architecture search (NAS) typically consists of three main steps:training a super-network, training and evaluating sampled deep neural networks(DNNs), and training the discovered DNN. Most of the existing efforts speed upsome steps at the cost of a significant slowdown of other steps or sacrificingthe support of non-differentiable search metrics. The unbalanced reduction inthe time spent per step limits the total search time reduction, and theinability to support non-differentiable search metrics limits the performanceof discovered DNNs.  In this paper, we present NetAdaptV2 with three innovations to better balancethe time spent for each step while supporting non-differentiable searchmetrics. First, we propose channel-level bypass connections that merge networkdepth and layer width into a single search dimension to reduce the time fortraining and evaluating sampled DNNs. Second, ordered dropout is proposed totrain multiple DNNs in a single forward-backward pass to decrease the time fortraining a super-network. Third, we propose the multi-layer coordinate descentoptimizer that considers the interplay of multiple layers in each iteration ofoptimization to improve the performance of discovered DNNs while supportingnon-differentiable search metrics. With these innovations, NetAdaptV2 reducesthe total search time by up to $5.8\times$ on ImageNet and $2.4\times$ on NYUDepth V2, respectively, and discovers DNNs with betteraccuracy-latency/accuracy-MAC trade-offs than state-of-the-art NAS works.Moreover, the discovered DNN outperforms NAS-discovered MobileNetV3 by 1.8%higher top-1 accuracy with the same latency. The project website ishttp://netadapt.mit.edu.

NetAdaptV2: Efficient Neural Architecture Search with Fast Super-Network Training and Architecture Optimization