Multi-Label Active Learning: Query Type Matters

Authors: Sheng-Jun Huang, Songcan Chen, Zhi-Hua Zhou

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on 20 benchmark datasets and a manually labeled real data validate that our approach not only achieves superior performance on classification, but also provides accurate ranking for relevant labels.
Researcher Affiliation Academia Sheng-Jun Huang1,3 and Songcan Chen1,3 and Zhi-Hua Zhou2,3 1College of Computer Science & Technology, Nanjing University of Aeronautics & Astronautics 2National Key Laboratory for Novel Software Technology, Nanjing University 3Collaborative Innovation Center of Novel Software Technology and Industrialization {huangsj, s.chen}@nuaa.edu.cn zhouzh@lamda.nju.edu.cn
Pseudocode No The paper describes the algorithm steps in paragraph form but does not include a formal pseudocode block or figure.
Open Source Code No The paper does not provide any statement or link regarding the release of open-source code for the described methodology.
Open Datasets Yes Experiments on 20 benchmark datasets and a manually labeled real data validate that our approach... The statistical information of these datasets are summarized in Table 1... MSRA is a multi-label dataset for image classification, and consists of 1868 images with 19 candidate labels.
Dataset Splits Yes For each experiment, we randomly divide the dataset into three parts: the test set with 50% examples, the initial labeled set with 5% examples and the unlabeled pool with the rest instances. Parameters are selected via leave-one-out cross validation on the initial set.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments.
Software Dependencies No The paper mentions using 'the multi-label learning algorithm proposed in the AUDI work [Huang and Zhou, 2013]' but does not specify any software names with version numbers (e.g., programming languages, libraries, or frameworks).
Experiment Setup Yes Parameters are selected via leave-one-out cross validation on the initial set... The algorithm iteratively samples data and employs stochastic gradient descent (SGD) to minimize the ranking error. At the t-th iteration of SGD, assuming the sampled triplet is (x, y, y), the model parameters can be updated according to: W t+1 0 = W t 0 γt i (wt yx wt yx ) (5) wt+1 y = wt y + γt i W t 0x (6) wt+1 y = wt y γt i W t 0x (7) where γt is the step size... After every 5 m queries, we evaluate the performance of the classification model on the test set.