Hierarchical Novelty Detection via Fine-Grained Evidence Allocation

Authors: Spandan Pyakurel, Qi Yu

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments conducted on real-world hierarchical datasets demonstrate the proposed model outperforms the strongest baselines and achieves the best HND performance.
Researcher Affiliation Academia 1Rochester Institute of Technology, Rochester, New York. Correspondence to: Qi Yu <qi.yu@rit.edu>.
Pseudocode Yes Algorithm 1 E-HND Training
Open Source Code Yes The source code can be accessed here: https://github.com/ritmininglab/EHND
Open Datasets Yes Tiny Imagenet (Le & Yang, 2015): It contains 200 classes each with 500 training, 50 validation, and 50 test images in each class, resulting in a total of 120k images.
Dataset Splits Yes Tiny Imagenet (Le & Yang, 2015): It contains 200 classes each with 500 training, 50 validation, and 50 test images in each class, resulting in a total of 120k images.
Hardware Specification Yes All the experiments are conducted using NVIDIA Ge Force RTX 3060 with 32GB memory.
Software Dependencies Yes The training algorithm is implemented in pytorch version: 1.13.0 and cuda version: 11.6.
Experiment Setup Yes We train the model using the full batch of Resnet-101 features with Adam optimizer and an initial learning rate of 10 2. We use the validation set to select the suitable hyperparameters β1 and β2. The validation set does not include samples from the novel classes. We use the set of (β1, β2) values of (65, 20), (30, 5), (20, 5) and (40, 5) for CUB, AWA2, Tiny Imagenet and Traffic respectively.