Poincaré Embeddings for Learning Hierarchical Representations
Authors: Maximillian Nickel, Douwe Kiela
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimentally, we show that our approach can provide high quality embeddings of large taxonomies both with and without missing data. Moreover, we show that embeddings trained on WORDNET provide state-of-the-art performance for lexical entailment. On collaboration networks, we also show that Poincaré embeddings are successful in predicting links in graphs where they outperform Euclidean embeddings, especially in low dimensions. |
| Researcher Affiliation | Industry | Maximilian Nickel Facebook AI Research maxn@fb.com Douwe Kiela Facebook AI Research dkiela@fb.com |
| Pseudocode | No | The paper provides mathematical derivations and descriptions of the update rule but no explicitly labeled 'Pseudocode' or 'Algorithm' block. |
| Open Source Code | No | The paper does not include a statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | We conduct experiments on the transitive closure of the WORDNET noun hierarchy [21] [...] We test this property on HYPERLEX [37], which is a gold standard resource [...] We performed our experiments on four commonly used social networks, i.e, ASTROPH, CONDMAT, GRQC, and HEPPH. |
| Dataset Splits | Yes | To test generalization performance, we split the data into a train, validation and test set by randomly holding out observed links. [...] For evaluation, we split each dataset randomly into train, validation, and test set. The hyperparameters r and t were tuned for each method on the validation set. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | First, we initialize all embeddings randomly from the uniform distribution U( 0.001, 0.001). [...] Second, we found that a good initial angular layout can be helpful to find good embeddings. For this reason, we train during an initial 'burn-in' phase with a reduced learning rate η/c. In our experiments, we set c = 10 and the duration of the burn-in to 10 epochs. [...] For training, we randomly sample 10 negative examples per positive example. [...] where r, t > 0 are hyperparameters. Here, r corresponds to the radius around each point u such that points within this radius are likely to have an edge with u. The parameter t specifies the steepness of the logistic function and influences both average clustering as well as the degree distribution [19]. |