Exploiting Conjugate Label Information for Multi-Instance Partial-Label Learning

Authors: Wei Tang, Weijia Zhang, Min-Ling Zhang

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results obtained from benchmark and real-world datasets demonstrate the superiority of the proposed ELIMIPL over existing MIPL algorithms and other well-established partial-label learning algorithms.
Researcher Affiliation Academia 1School of Computer Science and Engineering, Southeast University, Nanjing 210096, China 2Key Lab. of Computer Network and Information Integration (Southeast University), Mo E, China 3School of Information and Physical Sciences, The University of Newcastle, NSW 2308, Australia
Pseudocode Yes Algorithm 1 Training Procedure of ELIMIPL
Open Source Code Yes The code of ELIMIPL can be found at https://github.com/tangw-seu/ELIMIPL.
Open Datasets Yes We employ four benchmark MIPL datasets [Tang et al., 2024; Tang et al., 2023]: MNIST-MIPL, FMNIST-MIPL, Birdsong-MIPL, and SIVAL-MIPL, spanning diverse domains such as image analysis and biology [Le Cun et al., 1998; Xiao et al., 2017; Briggs et al., 2012; Settles et al., 2007]. The characteristics of the datasets are presented in Table 1
Dataset Splits No The paper mentions "train/test splits" but does not explicitly describe a separate validation split or how validation was performed if integrated (e.g., within the training split or via cross-validation).
Hardware Specification Yes We implement ELIMIPL using Py Torch and execute it on a single NVIDIA Tesla V100 GPU.
Software Dependencies No The paper mentions using "PyTorch" but does not specify its version number or versions of any other software dependencies needed to replicate the experiment.
Experiment Setup Yes We utilize the stochastic gradient descent (SGD) optimizer with a momentum value of 0.9 and a weight decay of 0.0001. The initial learning rate is selected from the set {0.01, 0.05} and accompanied by a cosine annealing technique. We set the number of epochs uniformly to 100 for all datasets. For the MNIST-MIPL and FMNIST-MIPL datasets, µ is set to 1 or 0.1, γ is chosen from {0.1, 0.5}, and the feature extraction network ψ1( ) is a two-layer convolutional neural network. For the remaining datasets, we set both µ and γ to 10, and ψ1( ) is an identity transformation. The feature transformation network ψ2( ) is implemented by a fully connected network, with the dimension l set to 512 for the CRC-MIPL dataset and 128 for the other datasets.