Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Progressive Purification for Instance-Dependent Partial Label Learning
Authors: Ning Xu, Biao Liu, Jiaqi Lv, Congyu Qiao, Xin Geng
ICML 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on the benchmark datasets and the realworld datasets validate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | 1School of Computer Science and Engineering, Southeast University, Nanjing, China. E-mail: EMAIL. 2RIKEN Center for Advanced Intelligence Project, Tokyo 103-0027, Japan. E-mail: EMAIL. |
| Pseudocode | Yes | Algorithm 1 POP Algorithm |
| Open Source Code | Yes | Source code is available at https://github.com/palm-ml/POP. |
| Open Datasets | Yes | We adopt five widely used benchmark datasets including MNIST (Le Cun et al., 1998), Kuzushiji-MNIST (Clanuwat et al., 2018), Fashion-MNIST (Xiao et al., 2017), CIFAR-10 (Krizhevsky & Hinton, 2009), CIFAR-100 (Krizhevsky & Hinton, 2009). In addition, seven real-world PLL datasets which are collected from different application domains are used, including Lost (Cour et al., 2011), Soccer Player (Zeng et al., 2013), Yahoo!News (Guillaumin et al., 2010) from automatic face naming, MSRCv2 (Liu & Dietterich, 2012) from object classification, Malagasy (Garrette & Baldridge, 2013) from POS tagging, Mirflickr (Huiskes & Lew, 2008) from web image classification, and Bird Song (Briggs et al., 2012) from bird song classification. |
| Dataset Splits | Yes | The hyper-parameters of the deep models are selected so as to maximize the accuracy on a validation set (10% of the training set). |
| Hardware Specification | No | The paper mentions models used (e.g., '5-layer Le Net', 'Wide-Res Net-28-2') and software ('Py Torch') but does not specify any hardware details such as GPU models, CPU types, or memory used for experiments. |
| Software Dependencies | No | The paper mentions 'Py Torch' as the implementation framework but does not provide specific version numbers for PyTorch or any other software dependencies. |
| Experiment Setup | Yes | We set e0 = 0.9, eend = 0.1 and es = 0.01. We run 5 trials on the benchmark datasets and the real-world PLL datasets. |