Coherent Hierarchical Multi-Label Classification Networks
Authors: Eleonora Giunchiglia, Thomas Lukasiewicz
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct an extensive experimental analysis showing the superior performance of C-HMCNN(h) when compared to state-of-the-art models. |
| Researcher Affiliation | Academia | Eleonora Giunchiglia Department of Computer Science University of Oxford, UK eleonora.giunchiglia@cs.ox.ac.uk Thomas Lukasiewicz Department of Computer Science University of Oxford, UK thomas.lukasiewicz@cs.ox.ac.uk |
| Pseudocode | No | The paper describes mathematical formulas and implementation logic, but it does not contain a clearly labeled 'Pseudocode' or 'Algorithm' block. |
| Open Source Code | Yes | Link: https://github.com/EGiunchiglia/C-HMCNN/ and the code os publicly available. |
| Open Datasets | Yes | We tested our model on 20 real-world datasets commonly used to compare HMC systems (see, e.g., [3, 23, 32, 33]): 16 are functional genomics datasets [9], 2 contain medical images [13], 1 contains images of microalgae [14], and 1 is a text categorization dataset [17].4. Links: https://dtai.cs.kuleuven.be/clus/hmcdatasets and http://kt.ijs.si/Dragi_Kocev/PhD/resources |
| Dataset Splits | Yes | The datasets consisted of 5000 (50/50 train test split) datapoints sampled from a uniform distribution over [0, 1]2. and The characteristics of these datasets are summarized in Table 1. Number of features (D), number of classes (n), and number of datapoints for each dataset split. |
| Hardware Specification | No | The paper mentions 'implemented on GPUs using standard libraries' and 'leveraging GPU architectures' but does not specify any particular GPU models, CPU types, or other hardware specifications. |
| Software Dependencies | No | The paper mentions 'Py Torch' as an example library but does not provide specific version numbers for any software dependencies, such as PyTorch, specific Adam optimizer implementations, or other libraries used for neural networks. |
| Experiment Setup | Yes | f and g were trained with binary cross-entropy loss using Adam optimization [16] for 20k epochs with learning rate 10 2 (β1 = 0.9, β2 = 0.999). and We built h as a feedforward neural network with two hidden layers and ReLU non-linearity. ...the loss was minimized using Adam optimizer with weight decay 10 5, and patience 20 (β1 = 0.9, β2 = 0.999). The dropout rate was set to 70% and the batch size to 4. |