Binary Classification with Confidence Difference

Authors: Wei Wang, Lei Feng, Yuchen Jiang, Gang Niu, Min-Ling Zhang, Masashi Sugiyama

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on benchmark data sets and a real-world recommender system data set validate the effectiveness of our proposed approaches in exploiting the supervision information of the confidence difference.
Researcher Affiliation Collaboration 1 The University of Tokyo, Chiba, Japan 2 RIKEN Center for Advanced Intelligence Project, Tokyo, Japan 3 Nanyang Technological University, Singapore 4 Alibaba Group, Beijing, China 5 Southeast University, Nanjing, China
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not include an explicit statement or link to the source code for the described methodology.
Open Datasets Yes We conducted experiments on benchmark data sets, including MNIST [61], Kuzushiji-MNIST [62], Fashion-MNIST [63], and CIFAR-10 [64]. In addition, four UCI data sets [65] were used, including Optdigits, USPS, Pendigits, and Letter.
Dataset Splits Yes MNIST [61]: It is composed of 60,000 training examples and 10,000 test examples.
Hardware Specification Yes All the experiments were conducted on NVIDIA Ge Force RTX 3090.
Software Dependencies No The paper mentions software like PyTorch and Adam optimizer but does not specify their version numbers.
Experiment Setup Yes The number of training epoches was set to 200 and we obtained the testing accuracy by averaging the results in the last 10 epoches. All the methods were implemented in Py Torch [69]. We used the Adam optimizer [70]. To ensure fair comparisons, We set the same hyperparameter values for all the compared approaches, where the details can be found in Appendix H. For MNIST, Kuzushiji-MNIST and Fashion-MNIST, the learning rate was set to 1e-3 and the weight decay was set to 1e-5. The batch size was set to 256 data pairs. For training the probabilistic classifier to generate confidence, the batch size was set to 256 and the epoch number was set to 10.