RECORD
One-shot Learning for Temporal Knowledge Graphs
Machine Learning · Submitted 23 Oct 2020
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our experiments show that the proposed algorithm outperforms the state of the art baselines for two well-studied benchmarks while achieving significantly better performance for sparse relations.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Most real-world knowledge graphs are characterized by a long-tail relation frequency distribution where a significant fraction of relations occurs only a handful of times. This observation has given rise to recent interest in low-shot learning methods that are able to generalize from only a few examples. The existing approaches, however, are tailored to static knowledge graphs and not easily generalized to temporal settings, where data scarcity poses even bigger problems, e.g., due to occurrence of new, previously unseen relations. We address this shortcoming by proposing a one-shot learning framework for link prediction in temporal knowledge graphs. Our proposed method employs a self-attention mechanism to effectively encode temporal interactions between entities, and a network to compute a similarity score between a given query and a (one-shot) example. Our experiments show that the proposed algorithm outperforms the state of the art baselines for two well-studied benchmarks while achieving significantly better performance for sparse relations.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Most real-world knowledge graphs are characterized by a long-tail relation frequency distribution where a significant fraction of relations occurs only a handful of times. | background 1.00 | 0.09 | 0.02 | asserted 1.00 |
| s1 | This observation has given rise to recent interest in low-shot learning methods that are able to generalize from only a few examples. | background 1.00 | 0.04 | 0.03 | asserted 0.98 |
| s2 | The existing approaches, however, are tailored to static knowledge graphs and not easily generalized to temporal settings, where data scarcity poses even bigger problems, e.g., due to occurrence of new, previously unseen relations. | gap 0.99 | 0.05 | 0.03 | asserted 0.98 |
| s3 | We address this shortcoming by proposing a one-shot learning framework for link prediction in temporal knowledge graphs. | contribution 1.00 | 0.82 | 0.11 | asserted 0.98 |
| s4 | Our proposed method employs a self-attention mechanism to effectively encode temporal interactions between entities, and a network to compute a similarity score between a given query and a (one-shot) example. | method 1.00 | 0.85 | 0.05 | asserted 1.00 |
| s5 | Our experiments show that the proposed algorithm outperforms the state of the art baselines for two well-studied benchmarks while achieving significantly better performance for sparse relations. | result 1.00 | 0.96 | 0.87 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Most real-world knowledge graphs are characterized by a long-tail relation frequency distribution where a significant fraction of relations occurs only a handful of times. This observation has given rise to recent interest in low-shot learning methods that are able to generalize from only a few examples. The existing approaches, however, are tailored to static knowledge graphs and not easily generalized to temporal settings, where data scarcity poses even bigger problems, e.g., due to occurrence of new, previously unseen relations. We address this shortcoming by proposing a one-shot learning framework for link prediction in temporal knowledge graphs. Our proposed method employs a self-attention mechanism to effectively encode temporal interactions between entities, and a network to compute a similarity score between a given query and a (one-shot) example. Our experiments show that the proposed algorithm outperforms the state of the art baselines for two well-studied benchmarks while achieving significantly better performance for sparse relations.
Human review
How the literature uses it
6 arXiv papers cite this in 6 sentences
Papers citing it · 4 sentences without a confident label
Cites it as background 2 papers
-
A few attempts have been proposed that combine meta learning with multi-hop reinforcement learning [ref] or sequential network [ref] to perform few-shot link prediction.
Subgraph-aware Few-Shot Inductive Link Prediction via Meta-Learning · Related Work · 0.91
-
Two such approaches that have gained significant attention are meta-learning [ref] and transfer learning [ref] .
Improved Region Proposal Network for Enhanced Few-Shot Object Detection · Related works · 1.00
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.