pith. sign in

arxiv: 1907.03143 · v1 · pith:445SV5VWnew · submitted 2019-07-06 · 💻 cs.LG · cs.AI· stat.ML

Diachronic Embedding for Temporal Knowledge Graph Completion

Rishab Goel , Seyed Mehran Kazemi , Marcus Brubaker , Pascal Poupart This is my paper

Pith reviewed 2026-05-25 01:29 UTC · model grok-4.3

classification 💻 cs.LG cs.AIstat.ML
keywords temporal knowledge graphsknowledge graph completiondiachronic embeddingsentity embeddingsSimplEKG embedding models
0
0 comments X

The pith

Equipping static knowledge graph models with a diachronic embedding function captures entity properties at any time and yields a fully expressive temporal completion model when combined with SimplE.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a diachronic embedding function that supplies time-specific characteristics for entities in knowledge graphs, allowing static embedding models to handle temporal facts without per-time-step parameters. This function is model-agnostic and learns from observed temporal relationships to represent how entities behave at arbitrary points in time. The authors prove that pairing the function with the SimplE static model produces a fully expressive system capable of representing any valid temporal fact. Experiments show the resulting models outperform existing temporal KG completion baselines.

Core claim

By adding a diachronic entity embedding function to static KG embedding models, temporal knowledge graph completion becomes possible in a way that contrasts with prior methods that rely only on static entity features; when combined with SimplE this produces a fully expressive model for the temporal case.

What carries the argument

The diachronic entity embedding function, which maps an entity and a timestamp to a time-specific vector representation used by an otherwise static scoring function.

If this is right

  • Any existing static KG embedding model can be extended to the temporal setting using the same function.
  • The SimplE combination can represent every possible temporal triple that is consistent with the data.
  • No separate embedding needs to be stored or learned for each distinct timestamp.
  • Temporal KG completion accuracy improves over methods that keep entity features fixed across time.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same function could be tested on dynamic graphs outside knowledge bases, such as evolving social or citation networks.
  • If the function generalizes, it may reduce the parameter count needed for modeling time-varying relations compared to time-specific architectures.
  • Scalability to very long time spans would depend on whether the function can extrapolate beyond the training time range.

Load-bearing premise

A single learnable function can produce accurate entity representations at any time from the available temporal facts without extra per-time parameters or supervision.

What would settle it

A dataset of temporal facts where entity properties change in a manner that no single parameterized function can fit across all observed times while still completing held-out facts correctly.

Figures

Figures reproduced from arXiv: 1907.03143 by Marcus Brubaker, Pascal Poupart, Rishab Goel, Seyed Mehran Kazemi.

Figure 1
Figure 1. Figure 1: (a) Test MRR of DE-SimplE on ICEWS14 as a function of [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
read the original abstract

Knowledge graphs (KGs) typically contain temporal facts indicating relationships among entities at different times. Due to their incompleteness, several approaches have been proposed to infer new facts for a KG based on the existing ones-a problem known as KG completion. KG embedding approaches have proved effective for KG completion, however, they have been developed mostly for static KGs. Developing temporal KG embedding models is an increasingly important problem. In this paper, we build novel models for temporal KG completion through equipping static models with a diachronic entity embedding function which provides the characteristics of entities at any point in time. This is in contrast to the existing temporal KG embedding approaches where only static entity features are provided. The proposed embedding function is model-agnostic and can be potentially combined with any static model. We prove that combining it with SimplE, a recent model for static KG embedding, results in a fully expressive model for temporal KG completion. Our experiments indicate the superiority of our proposal compared to existing baselines.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript proposes equipping static knowledge graph embedding models with a parameterized diachronic entity embedding function to handle temporal facts. It claims this function is model-agnostic and proves that its combination with SimplE yields a fully expressive model for temporal KG completion (able to realize arbitrary truth assignments to (s,r,o,t) quadruples). Experiments are reported to show superiority over existing temporal KG completion baselines.

Significance. If the full-expressiveness proof holds under the stated parameterization and the experimental gains are robust, the work supplies a general mechanism for converting static models into temporal ones while preserving expressivity guarantees. This could serve as a template for other static embeddings and reduce the need for per-time-step parameters.

major comments (1)
  1. [Abstract / proof of full expressiveness] Abstract and proof of full expressiveness: the claim that the diachronic function + SimplE is fully expressive assumes the function can realize independent temporal trajectories for each entity. The manuscript must clarify whether the function parameters θ are global (shared across all entities) or permitted to grow with the number of entities; a globally shared parameterization forces identical functional forms of time dependence and precludes arbitrary per-entity temporal patterns, undermining the expressiveness result.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and for identifying the need for explicit clarification on the parameterization of the diachronic embedding function. The concern is valid and we address it directly below.

read point-by-point responses

  1. Referee: [Abstract / proof of full expressiveness] Abstract and proof of full expressiveness: the claim that the diachronic function + SimplE is fully expressive assumes the function can realize independent temporal trajectories for each entity. The manuscript must clarify whether the function parameters θ are global (shared across all entities) or permitted to grow with the number of entities; a globally shared parameterization forces identical functional forms of time dependence and precludes arbitrary per-entity temporal patterns, undermining the expressiveness result.

    Authors: We agree that the parameterization must be clarified. In the proposed model the diachronic embedding function is instantiated separately for each entity: for entity e the parameters θ_e (e.g., the coefficients of the chosen basis functions) are learned independently and therefore grow linearly with the number of entities. This per-entity parameterization is what permits arbitrary temporal trajectories and is the assumption underlying the full-expressiveness proof for SimplE. A globally shared θ would indeed collapse the model to a single functional form and would invalidate the expressiveness claim. We will revise the manuscript (both the abstract and the proof section) to state explicitly that θ is entity-specific and to include a short remark confirming that the proof relies on this per-entity parameterization. revision: yes

Circularity Check

0 steps flagged

Minor self-citation to SimplE; central full-expressiveness proof is independent

full rationale

The paper introduces a parameterized diachronic embedding function that is model-agnostic and proves that its combination with SimplE produces a fully expressive temporal model. This proof is developed directly in the present work from the definitions of the diachronic function and the static SimplE model; it does not reduce any claimed result to a fitted quantity defined by the same data or to a self-referential construction. The citation to SimplE (prior work by an overlapping author) is used as a building block but is not load-bearing for uniqueness or for the temporal extension itself. No other patterns (self-definitional, fitted-input-as-prediction, ansatz smuggling, or renaming) appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Abstract-only review; no explicit free parameters, axioms, or invented entities are detailed beyond the introduction of the diachronic embedding function itself.

invented entities (1)
  • diachronic entity embedding function no independent evidence
    purpose: to supply time-varying characteristics of entities for any point in time
    Introduced as the core new component that distinguishes the proposal from static-feature temporal models.

pith-pipeline@v0.9.0 · 5710 in / 1057 out tokens · 24470 ms · 2026-05-25T01:29:43.164624+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Temporal Graph Networks for Deep Learning on Dynamic Graphs

    cs.LG 2020-06 unverdicted novelty 7.0

    Temporal Graph Networks combine memory modules and graph operators to learn on dynamic graphs as timed event sequences, outperforming prior methods on transductive and inductive tasks while unifying earlier models as ...

Reference graph

Works this paper leans on

65 extracted references · 65 canonical work pages · cited by 1 Pith paper · 7 internal anchors

  1. [1]

    Hypernetwork knowledge graph embeddings

    Ivana Balazevic, Carl Allen, and Timothy M Hospedales. Hypernetwork knowledge graph embeddings. arXiv preprint arXiv:1808.07018, 2018

    work page arXiv 2018

  2. [2]

    Tucker: Tensor factorization for knowledge graph completion

    Ivana Balaževi´c, Carl Allen, and Timothy M Hospedales. Tucker: Tensor factorization for knowledge graph completion. arXiv preprint arXiv:1901.09590, 2019

    work page arXiv 1901

  3. [3]

    Dynamic word embeddings

    Robert Bamler and Stephan Mandt. Dynamic word embeddings. In ICML, pages 380–389, 2017

    work page 2017

  4. [4]

    Translating embeddings for modeling multi-relational data

    Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Oksana Yakhnenko. Translating embeddings for modeling multi-relational data. In NeurIPS, pages 2787–2795, 2013

    work page 2013

  5. [5]

    Icews coded event data

    Elizabeth Boschee, Jennifer Lautenschlager, Sean O’Brien, Steve Shellman, James Starz, and Michael Ward. Icews coded event data. Harvard Dataverse, 12, 2015

    work page 2015

  6. [6]

    Negation without negation in probabilistic logic programming

    David Buchman and David Poole. Negation without negation in probabilistic logic programming. In KR, 2016

    work page 2016

  7. [7]

    Introduction to the Theory of Fourier’s Series and Integrals

    Horatio Scott Carslaw. Introduction to the Theory of Fourier’s Series and Integrals. Macmillan, 1921

    work page 1921

  8. [8]

    Rule based temporal inference

    Melisachew Wudage Chekol and Heiner Stuckenschmidt. Rule based temporal inference. In ICLP. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2018

    work page 2018

  9. [9]

    Marrying uncertainty and time in knowledge graphs

    Melisachew Wudage Chekol, Giuseppe Pirrò, Joerg Schoenfisch, and Heiner Stuckenschmidt. Marrying uncertainty and time in knowledge graphs. In AAAI, 2017

    work page 2017

  10. [10]

    Hyte: Hyperplane-based temporally aware knowledge graph embedding

    Shib Sankar Dasgupta, Swayambhu Nath Ray, and Partha Talukdar. Hyte: Hyperplane-based temporally aware knowledge graph embedding. In EMNLP, pages 2001–2011, 2018

    work page 2001

  11. [11]

    Problog: A probabilistic prolog and its application in link discovery

    Luc De Raedt, Angelika Kimmig, and Hannu Toivonen. Problog: A probabilistic prolog and its application in link discovery. In IJCAI, volume 7, pages 2462–2467. Hyderabad, 2007

    work page 2007

  12. [12]

    Convolutional 2d knowledge graph embeddings

    Tim Dettmers, Pasquale Minervini, Pontus Stenetorp, and Sebastian Riedel. Convolutional 2d knowledge graph embeddings. In AAAI, 2018

    work page 2018

  13. [13]

    Knowledge vault: A web-scale approach to probabilistic knowledge fusion

    Xin Dong, Evgeniy Gabrilovich, Geremy Heitz, Wilko Horn, Ni Lao, Kevin Murphy, Thomas Strohmann, Shaohua Sun, and Wei Zhang. Knowledge vault: A web-scale approach to probabilistic knowledge fusion. In ACM SIGKDD, pages 601–610. ACM, 2014

    work page 2014

  14. [14]

    A temporal-probabilistic database model for information extraction

    Maximilian Dylla, Iris Miliaraki, and Martin Theobald. A temporal-probabilistic database model for information extraction. Proceedings of the VLDB Endowment, 6(14):1810–1821, 2013

    work page 2013

  15. [15]

    Improved knowledge graph embedding using background taxonomic information

    Bahare Fatemi, Siamak Ravanbakhsh, and David Poole. Improved knowledge graph embedding using background taxonomic information. In AAAI, 2019

    work page 2019

  16. [16]

    Learning Sequence Encoders for Temporal Knowledge Graph Completion

    Alberto García-Durán, Sebastijan Dumanˇci´c, and Mathias Niepert. Learning sequence encoders for temporal knowledge graph completion. arXiv preprint arXiv:1809.03202, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  17. [17]

    Taming the waves: sine as activation function in deep neural networks

    Tuomas Virtanen Giambattista Parascandolo, Heikki Huttunen. Taming the waves: sine as activation function in deep neural networks. 2017

    work page 2017

  18. [18]

    Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change

    William L Hamilton, Jure Leskovec, and Dan Jurafsky. Diachronic word embeddings reveal statistical laws of semantic change. arXiv preprint arXiv:1605.09096, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  19. [19]

    The expression of a tensor or a polyadic as a sum of products

    Frank L Hitchcock. The expression of a tensor or a polyadic as a sum of products. Journal of Mathematics and Physics, 6(1-4):164–189, 1927

    work page 1927

  20. [20]

    Applying Markov logic for debugging probabilistic temporal knowledge bases

    Jakob Huber, Christian Meilicke, and Heiner Stuckenschmidt. Applying Markov logic for debugging probabilistic temporal knowledge bases. In AKBC, 2014

    work page 2014

  21. [21]

    Towards time-aware knowledge graph completion

    Tingsong Jiang, Tianyu Liu, Tao Ge, Lei Sha, Baobao Chang, Sujian Li, and Zhifang Sui. Towards time-aware knowledge graph completion. In COLING, pages 1715–1724, 2016. 9

    work page 2016

  22. [22]

    Knowledge Base Completion: Baselines Strike Back

    Ondrej Bajgar Kadlec, Rudolf and Jan Kleindienst. Knowledge base completion: Baselines strike back. arXiv preprint arXiv:1705.10744, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  23. [23]

    Bridging weighted rules and graph random walks for statistical relational models

    Seyed Mehran Kazemi and David Poole. Bridging weighted rules and graph random walks for statistical relational models. Frontiers in Robotics and AI, 5:8, 2018

    work page 2018

  24. [24]

    RelNN: A deep neural model for relational learning

    Seyed Mehran Kazemi and David Poole. RelNN: A deep neural model for relational learning. In AAAI, 2018

    work page 2018

  25. [25]

    SimplE embedding for link prediction in knowledge graphs

    Seyed Mehran Kazemi and David Poole. SimplE embedding for link prediction in knowledge graphs. In NeurIPS, pages 4289–4300, 2018

    work page 2018

  26. [26]

    Relational logistic regression

    Seyed Mehran Kazemi, David Buchman, Kristian Kersting, Sriraam Natarajan, and David Poole. Relational logistic regression. In KR, 2014

    work page 2014

  27. [27]

    Relational representation learning for dynamic (knowledge) graphs: A survey

    Seyed Mehran Kazemi, Rishab Goel, Kshitij Jain, Ivan Kobyzev, Akshay Sethi, Peter Forsyth, and Pascal Poupart. Relational representation learning for dynamic (knowledge) graphs: A survey. arXiv preprint arXiv:1905.11485, 2019

    work page arXiv 1905

  28. [28]

    Temporal Analysis of Language through Neural Language Models

    Yoon Kim, Yi-I Chiu, Kentaro Hanaki, Darshan Hegde, and Slav Petrov. Temporal analysis of language through neural language models. arXiv preprint arXiv:1405.3515, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  29. [29]

    A short introduction to probabilistic soft logic

    Angelika Kimmig, Stephen H Bach, Matthias Broecheler, Bert Huang, and Lise Getoor. A short introduction to probabilistic soft logic. In NIPS Workshop on probabilistic programming: Foundations and applications, volume 1, page 3, 2012

    work page 2012

  30. [30]

    Adam: A Method for Stochastic Optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  31. [31]

    Introduction to statistical relational learning

    Daphne Koller, Nir Friedman, Sašo Džeroski, Charles Sutton, Andrew McCallum, Avi Pfeffer, Pieter Abbeel, Ming-Fai Wong, David Heckerman, Chris Meek, et al. Introduction to statistical relational learning. MIT press, 2007

    work page 2007

  32. [32]

    Statistically significant detection of linguistic change

    Vivek Kulkarni, Rami Al-Rfou, Bryan Perozzi, and Steven Skiena. Statistically significant detection of linguistic change. In WWW, pages 625–635, 2015

    work page 2015

  33. [33]

    Learning Dynamic Embeddings from Temporal Interactions

    Srijan Kumar, Xikun Zhang, and Jure Leskovec. Learning dynamic embedding from temporal interaction networks. arXiv preprint arXiv:1812.02289, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  34. [34]

    Canonical tensor decomposition for knowledge base completion

    Timothée Lacroix, Nicolas Usunier, and Guillaume Obozinski. Canonical tensor decomposition for knowledge base completion. In ICML, 2018

    work page 2018

  35. [35]

    Relational retrieval using a combination of path-constrained random walks

    Ni Lao and William W Cohen. Relational retrieval using a combination of path-constrained random walks. Machine learning, 81(1):53–67, 2010

    work page 2010

  36. [36]

    Random walk inference and learning in a large scale knowledge base

    Ni Lao, Tom Mitchell, and William W Cohen. Random walk inference and learning in a large scale knowledge base. In EMNLP, pages 529–539, 2011

    work page 2011

  37. [37]

    Nonlinear signal processing using neural networks: Prediction and system modelling

    Alan Lapedes and Robert Farber. Nonlinear signal processing using neural networks: Prediction and system modelling. Technical report, 1987

    work page 1987

  38. [38]

    Gdelt: Global data on events, location, and tone, 1979–

    Kalev Leetaru and Philip A Schrodt. Gdelt: Global data on events, location, and tone, 1979–

    work page 1979

  39. [39]

    Citeseer, 2013

    In ISA annual convention, volume 2, pages 1–49. Citeseer, 2013

    work page 2013

  40. [40]

    Learning entity and relation embeddings for knowledge graph completion

    Yankai Lin, Zhiyuan Liu, Maosong Sun, Yang Liu, and Xuan Zhu. Learning entity and relation embeddings for knowledge graph completion. In AAAI, pages 2181–2187, 2015

    work page 2015

  41. [41]

    Analogical inference for multi-relational embed- dings

    Hanxiao Liu, Yuexin Wu, and Yiming Yang. Analogical inference for multi-relational embed- dings. In ICML, pages 2168–2178, 2017

    work page 2017

  42. [42]

    Embedding models for episodic knowledge graphs

    Yunpu Ma, V olker Tresp, and Erik A Daxberger. Embedding models for episodic knowledge graphs. Journal of Web Semantics, 2018

    work page 2018

  43. [43]

    Regularizing knowledge graph embeddings via equivalence and inversion axioms

    Pasquale Minervini, Luca Costabello, Emir Muñoz, Vít Nováˇcek, and Pierre-Yves Vandenbuss- che. Regularizing knowledge graph embeddings via equivalence and inversion axioms. In ECML PKDD, pages 668–683. Springer, 2017. 10

    work page 2017

  44. [44]

    Stranse: a novel embedding model of entities and relationships in knowledge bases

    Dat Quoc Nguyen, Kairit Sirts, Lizhen Qu, and Mark Johnson. Stranse: a novel embedding model of entities and relationships in knowledge bases. In NAACL-HLT, 2016

    work page 2016

  45. [45]

    An overview of embedding models of entities and relationships for knowl- edge base completion

    Dat Quoc Nguyen. An overview of embedding models of entities and relationships for knowl- edge base completion. arXiv preprint arXiv:1703.08098, 2017

    work page arXiv 2017

  46. [46]

    A three-way model for collective learning on multi-relational data

    Maximilian Nickel, V olker Tresp, and Hans-Peter Kriegel. A three-way model for collective learning on multi-relational data. In ICML, volume 11, pages 809–816, 2011

    work page 2011

  47. [47]

    A review of relational machine learning for knowledge graphs

    Maximilian Nickel, Kevin Murphy, V olker Tresp, and Evgeniy Gabrilovich. A review of relational machine learning for knowledge graphs. Proceedings of the IEEE, 104(1):11–33, 2016

    work page 2016

  48. [48]

    Holographic embeddings of knowledge graphs

    Maximilian Nickel, Lorenzo Rosasco, and Tomaso Poggio. Holographic embeddings of knowledge graphs. In AAAI, 2016

    work page 2016

  49. [49]

    Slice normalized dynamic markov logic networks

    Tivadar Papai, Henry Kautz, and Daniel Stefankovic. Slice normalized dynamic markov logic networks. In NeurIPS, pages 1907–1915, 2012

    work page 1907

  50. [50]

    Automatic differentiation in pytorch

    Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Alban Desmaison, Luca Antiga, and Adam Lerer. Automatic differentiation in pytorch. In NIPS-W, 2017

    work page 2017

  51. [51]

    Statistical relational artificial intelligence: Logic, probability, and computation

    Luc De Raedt, Kristian Kersting, Sriraam Natarajan, and David Poole. Statistical relational artificial intelligence: Logic, probability, and computation. Synthesis Lectures on Artificial Intelligence and Machine Learning, 10(2):1–189, 2016

    work page 2016

  52. [52]

    Markov logic networks

    Matthew Richardson and Pedro Domingos. Markov logic networks. Machine learning, 62(1- 2):107–136, 2006

    work page 2006

  53. [53]

    Recognizing multi-agent activities from gps data

    Adam Sadilek and Henry Kautz. Recognizing multi-agent activities from gps data. In AAAI, 2010

    work page 2010

  54. [54]

    Reasoning with neural tensor networks for knowledge base completion

    Richard Socher, Danqi Chen, Christopher D Manning, and Andrew Ng. Reasoning with neural tensor networks for knowledge base completion. In AAAI, pages 926–934, 2013

    work page 2013

  55. [55]

    Lifted Relational Neural Networks

    Gustav Sourek, V ojtech Aschenbrenner, Filip Zelezny, and Ondrej Kuzelka. Lifted relational neural networks. arXiv preprint arXiv:1508.05128, 2015

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  56. [56]

    RotatE: Knowledge graph embed- ding by relational rotation in complex space

    Zhiqing Sun, Zhi-Hong Deng, Jian-Yun Nie, and Jian Tang. RotatE: Knowledge graph embed- ding by relational rotation in complex space. In ICLR, 2019

    work page 2019

  57. [57]

    Know-evolve: Deep temporal reasoning for dynamic knowledge graphs

    Rakshit Trivedi, Hanjun Dai, Yichen Wang, and Le Song. Know-evolve: Deep temporal reasoning for dynamic knowledge graphs. In ICML, pages 3462–3471, 2017

    work page 2017

  58. [58]

    DyRep: Learning representations over dynamic graphs

    Rakshit Trivedi, Mehrdad Farajtabar, Prasenjeet Biswal, and Hongyuan Zha. DyRep: Learning representations over dynamic graphs. In ICLR, 2019

    work page 2019

  59. [59]

    Complex embeddings for simple link prediction

    Théo Trouillon, Johannes Welbl, Sebastian Riedel, Éric Gaussier, and Guillaume Bouchard. Complex embeddings for simple link prediction. In ICML, pages 2071–2080, 2016

    work page 2071

  60. [60]

    Knowledge graph completion via complex tensor factorization

    Théo Trouillon, Christopher R Dance, Éric Gaussier, Johannes Welbl, Sebastian Riedel, and Guillaume Bouchard. Knowledge graph completion via complex tensor factorization. JMLR, 18(1):4735–4772, 2017

    work page 2017

  61. [61]

    Some mathematical notes on three-mode factor analysis

    Ledyard R Tucker. Some mathematical notes on three-mode factor analysis. Psychometrika, 31(3):279–311, 1966

    work page 1966

  62. [62]

    Knowledge graph embedding by translating on hyperplanes

    Zhen Wang, Jianwen Zhang, Jianlin Feng, and Zheng Chen. Knowledge graph embedding by translating on hyperplanes. In AAAI, 2014

    work page 2014

  63. [63]

    Knowledge graph embedding: A survey of approaches and applications

    Quan Wang, Zhendong Mao, Bin Wang, and Li Guo. Knowledge graph embedding: A survey of approaches and applications. IEEE TKDE, 29(12):2724–2743, 2017

    work page 2017

  64. [64]

    How powerful are graph neural networks? In ICLR, 2019

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? In ICLR, 2019

    work page 2019

  65. [65]

    Embedding entities and relations for learning and inference in knowledge bases

    Bishan Yang, Wen-tau Yih, Xiaodong He, Jianfeng Gao, and Li Deng. Embedding entities and relations for learning and inference in knowledge bases. ICLR, 2015. 11 A Proof of Theorems and Propositions Theorem 1. DE-SimplE is fully expressive for temporal knowledge graph completion. Proof. For every entity vi∈V , let DEEMB(vi, t) = ( ⃗zzzt vi , ⃗zzzt vi ) whe...

    work page 2015