Active Discriminative Text Representation Learning
Authors: Ye Zhang, Matthew Lease, Byron Wallace
AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Empirical results show that our method outperforms baseline AL approaches on both sentence and document classification tasks. We also show that, as expected, the method quickly learns discriminative word embeddings. |
| Researcher Affiliation | Academia | Ye Zhang Department of Computer Science University of Texas at Austin yezhang@utexas.edu Matthew Lease School of Information University of Texas at Austin ml@utexas.edu Byron C. Wallace College of Computer & Information Science Northeastern University byron@ccs.neu.edu |
| Pseudocode | No | Not found. The paper describes the models and algorithms using text and mathematical equations, but it does not include any structured pseudocode blocks or sections labeled as "Algorithm". |
| Open Source Code | No | Not found. The paper does not provide any explicit statements or links about making the source code for their proposed methodology publicly available. |
| Open Datasets | Yes | CR: positive / negative product reviews (Hu and Liu 2004).2 MR: positive / negative movie reviews (Pang and Lee 2005). Subj: subjective / objective sentences (Pang and Lee 2004).3 ... Mu R: Music reviews (Blitzer et al. 2007).5 |
| Dataset Splits | Yes | We performed 20 rounds of batch active learning. At the outset, we provided all learners with the same 25 instances (sampled i.i.d. at random). In subsequent rounds, each learner was allowed to select 25 instances from U according to their respective querying strategies. These examples were added to L, and the models were retrained. ... For all but one dataset we repeated this entire AL process 10 times, using test sets generated via 10-fold CV. |
| Hardware Specification | No | Not found. The paper does not specify any details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments. |
| Software Dependencies | No | Not found. The paper mentions using "Adadelta (Zeiler 2012)" as the optimizer, but it does not provide version numbers for Adadelta or any other software libraries or frameworks used in the implementation. |
| Experiment Setup | Yes | We used three filter heights (3, 4, 5). For sentence and document classification tasks, we used 50 and 100 filters of each size, respectively. ... We performed 20 rounds of batch active learning. At the outset, we provided all learners with the same 25 instances (sampled i.i.d. at random). ... For EGL-Entropy Beta, we fixed α = 2 and initialized β = 2 as well, which implies a roughly equal weight on embedding and uncertainty scores. We then decreased βt linearly with iterations t. ... We estimated parameters by Adadelta (Zeiler 2012), tuning E in back-propagation to induce discriminative embeddings. |