Rethink DARTS Search Space and Renovate a New Benchmark

Authors: Jiuling Zhang, Zhiming Ding

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We re-implement twelve baselines and evaluate them across twelve conditions by combining two underexpolored influential factors: transductive robustness and discretization policy, to reasonably construct a benchmark upon multi-condition evaluation.
Researcher Affiliation Academia 1University of Chinese Academy of Sciences, Beijing, China 2Institute of Software, Chinese Academy of Sciences, Beijing, China.
Pseudocode No The paper does not contain any sections explicitly labeled 'Pseudocode' or 'Algorithm', nor does it present any structured, code-like blocks describing a procedure.
Open Source Code Yes https://github.com/chaoji90/LHD
Open Datasets Yes Our benchmark is evaluated on the most commonly used standard benchmark datasets (CIFAR-10, CIFAR-100, SVHN)
Dataset Splits Yes architecture parameters α and operation weights ω are alternately optimized on validation set and training set respectively through a bilevel optimization objective.
Hardware Specification Yes 5.2h 3.1h on RTX 3090, like-for-like comparison after aligning all other conditions).
Software Dependencies Yes Software version for search and evaluation of the benchmark: torch 1.9, cuda 11.1, cudnn 8.2, driver version 460.67. But we also test the search and evaluation codes and verify the empirical memory overhead on more recent version: torch 1.10, cuda 11.3, cudnn 8.3 and driver 495.44.
Experiment Setup Yes Table 8. Hyperparameter settings of baselines in search.