Online Learning in Variable Feature Spaces under Incomplete Supervision
Authors: Yi He, Xu Yuan, Sheng Chen, Xindong Wu4106-4114
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | This paper explores a new online learning problem where the input sequence lives in an over-time varying feature space and the ground-truth label of any input point is given only occasionally, making online learners less restrictive and more applicable. Theoretical analysis and empirical evidence substantiate the viability and effectiveness of our proposal. This section aims to experimentally validate whether our solution is viable and effective to the OVSIS problem. Our evaluations are conducted on 10 datasets, including 8 from the UCI repository (Dua and Karra Taniskidou 2017) and 2 from the datasets of IMDB (Maas et al. 2011) and CCYS (He et al. 2021). Table 2 presents the results of performance comparison in terms of classification accuracy. |
| Researcher Affiliation | Collaboration | 1 Center for Advanced Computer Studies (CACS), University of Louisiana at Lafayette, USA 2 Key Laboratory of Knowledge Engineering with Big Data, Hefei University of Technology, China 3 Mininglamp Academy of Sciences, Mininglamp Technology, China |
| Pseudocode | Yes | Algorithm 1: The AGDES Algorithm |
| Open Source Code | No | The paper does not provide an explicit statement about releasing the source code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | Our evaluations are conducted on 10 datasets, including 8 from the UCI repository (Dua and Karra Taniskidou 2017) and 2 from the datasets of IMDB (Maas et al. 2011) and CCYS (He et al. 2021). The UCI repository, IMDB, and CCYS datasets are well-known public datasets, and proper citations are provided for each. |
| Dataset Splits | No | The paper does not explicitly state specific training, validation, or test dataset splits (e.g., percentages or sample counts). It only mentions that experiments are repeated 5 times with random permutation. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to run the experiments (e.g., CPU, GPU models, or cloud computing specifications). |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers that would be required to reproduce the experiments. |
| Experiment Setup | Yes | To simulate label scarceness, we randomly mask 70% labels (i.e., pl = 0.3). The objective Eq. (7) is governed by two parameters λ1 and λ2 that need to be determined in an ad hoc way. We investigate the impact of parameter values on our approach. The parameter λ1 decides how slack the reconstruction errors are tolerated. The parameter λ2 controls how strongly the learner preserves the manifold structure. We grid search λ1 in {1e 3, 5e 3, . . . , .075} and λ2 in {1.5e 5, 5e 5, . . . , .085}, and present corresponding accuracy of AGDES in Figure 3. |