Exploring Diverse Representations for Open Set Recognition

Authors: Yu Wang, Junxian Mu, Pengfei Zhu, Qinghua Hu

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments are conducted on standard and OSR largescale benchmarks. Results show that the proposed discriminative method can outperform existing generative models by up to 9.5% on AUROC and achieve new state-of-the-art performance with little computational cost.
Researcher Affiliation Academia Yu Wang, Junxian Mu, Pengfei Zhu*, Qinghua Hu 1College of Intelligence and Computing, Tianjin University wang.yu@tju.edu.cn, jxmu@tju.edu.cn, zhupengfei@tju.edu.cn, huqinghua@tju.edu.cn
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Code is available at https://github.com/Vanixxz/MEDAF.
Open Datasets Yes we conducted the experiments on five image datasets, including CIFAR10 (Krizhevsky 2009), CIFAR+10, CIFAR+50, SVHN (Netzer et al. 2011), and Tiny-Image Net (Pouransari and Ghili 2014).
Dataset Splits Yes To avoid unfair comparisons arising from different splits, we adopted unified split information with (Moon et al. 2022; Neal et al. 2018) and can be found in the supplementary.
Hardware Specification No The paper does not provide specific hardware details such as GPU models, CPU types, or memory specifications used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python version, library versions like PyTorch or TensorFlow) beyond mentioning the use of Res Net18 and SGD optimizer.
Experiment Setup Yes In terms of optimization, we used an SGD optimizer with a momentum value of 0.9 and set the initial learning rate to 0.1 with a fixed batch size of 128 for 150 epochs.