GHOST: Gaussian Hypothesis Open-Set Technique
Authors: Ryan Rabinowitz, Steve Cruz, Manuel Günther, Terrance E. Boult
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate GHOST across multiple Image Net-1K pre-trained deep networks and test it with four different unknown datasets. Using standard metrics such as AUOSCR, AUROC and FPR95, we achieve statistically significant improvements, advancing the state-of-the-art in large-scale OSR. |
| Researcher Affiliation | Academia | 1Vision and Security Technology Lab, University of Colorado Colorado Springs, 2Computer Vision Research Lab, University of Notre Dame, 3Department of Informatics, University of Zurich EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes the GHOST approach, training, and inference process in descriptive text and diagrams (Figure 2), but it does not include a formal pseudocode block or algorithm section. |
| Open Source Code | Yes | Source code is provided online. Code Repository https://github.com/Vastlab/GHOST |
| Open Datasets | Yes | We utilize a recent purpose-built OOD dataset called No Image Net Class Objects (NINCO) (Bitterwolf, Mueller, and Hein 2023)... Additionally, we use the Image Net21K-P Open-Set splits (Easy & Hard) proposed by Vaze et al. (2022)... We also include Open Image-O (Wang et al. 2022)... Further details on each dataset and comparisons on additional datasets such as Places (Zhou et al. 2017), SUN (Xiao et al. 2010), and Textures (Cimpoi et al. 2014) are provided in the supplemental. |
| Dataset Splits | Yes | In particular, we use Image Net-1K (Russakovsky et al. 2015) pre-trained networks and the validation set as our test set for knowns. |
| Hardware Specification | No | The paper discusses the architectures used (MAE-H and Conv Ne Xt V2-H) which were pre-trained, but it does not provide specific details about the hardware (e.g., GPU models, CPU models, or memory) used for running the experiments described in this paper. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python version, PyTorch version, specific library versions). It mentions using deep neural networks but no software stack details. |
| Experiment Setup | No | While the paper states that GHOST is hyperparameter-free, it does not provide specific experimental setup details such as concrete hyperparameter values (e.g., learning rates, batch sizes, epochs) for the baseline methods or the general training/evaluation configurations used in their experiments. |