---
_id: '27381'
abstract:
- lang: eng
  text: Graph neural networks (GNNs) have been successfully applied in many structured
    data domains, with applications ranging from molecular property prediction to
    the analysis of social networks. Motivated by the broad applicability of GNNs,
    we propose the family of so-called RankGNNs, a combination of neural Learning
    to Rank (LtR) methods and GNNs. RankGNNs are trained with a set of pair-wise preferences
    between graphs, suggesting that one of them is preferred over the other. One practical
    application of this problem is drug screening, where an expert wants to find the
    most promising molecules in a large collection of drug candidates. We empirically
    demonstrate that our proposed pair-wise RankGNN approach either significantly
    outperforms or at least matches the ranking performance of the naive point-wise
    baseline approach, in which the LtR problem is solved via GNN-based graph regression.
author:
- first_name: Clemens
  full_name: Damke, Clemens
  id: '48192'
  last_name: Damke
  orcid: 0000-0002-0455-0048
- first_name: Eyke
  full_name: Hüllermeier, Eyke
  id: '48129'
  last_name: Hüllermeier
citation:
  ama: 'Damke C, Hüllermeier E. Ranking Structured Objects with Graph Neural Networks.
    In: Soares C, Torgo L, eds. <i>Proceedings of The 24th International Conference
    on Discovery Science (DS 2021)</i>. Vol 12986. Lecture Notes in Computer Science.
    Springer; 2021:166-180. doi:<a href="https://doi.org/10.1007/978-3-030-88942-5">10.1007/978-3-030-88942-5</a>'
  apa: Damke, C., &#38; Hüllermeier, E. (2021). Ranking Structured Objects with Graph
    Neural Networks. In C. Soares &#38; L. Torgo (Eds.), <i>Proceedings of The 24th
    International Conference on Discovery Science (DS 2021)</i> (Vol. 12986, pp. 166–180).
    Springer. <a href="https://doi.org/10.1007/978-3-030-88942-5">https://doi.org/10.1007/978-3-030-88942-5</a>
  bibtex: '@inproceedings{Damke_Hüllermeier_2021, series={Lecture Notes in Computer
    Science}, title={Ranking Structured Objects with Graph Neural Networks}, volume={12986},
    DOI={<a href="https://doi.org/10.1007/978-3-030-88942-5">10.1007/978-3-030-88942-5</a>},
    booktitle={Proceedings of The 24th International Conference on Discovery Science
    (DS 2021)}, publisher={Springer}, author={Damke, Clemens and Hüllermeier, Eyke},
    editor={Soares, Carlos and Torgo, Luis}, year={2021}, pages={166–180}, collection={Lecture
    Notes in Computer Science} }'
  chicago: Damke, Clemens, and Eyke Hüllermeier. “Ranking Structured Objects with
    Graph Neural Networks.” In <i>Proceedings of The 24th International Conference
    on Discovery Science (DS 2021)</i>, edited by Carlos Soares and Luis Torgo, 12986:166–80.
    Lecture Notes in Computer Science. Springer, 2021. <a href="https://doi.org/10.1007/978-3-030-88942-5">https://doi.org/10.1007/978-3-030-88942-5</a>.
  ieee: 'C. Damke and E. Hüllermeier, “Ranking Structured Objects with Graph Neural
    Networks,” in <i>Proceedings of The 24th International Conference on Discovery
    Science (DS 2021)</i>, Halifax, Canada, 2021, vol. 12986, pp. 166–180, doi: <a
    href="https://doi.org/10.1007/978-3-030-88942-5">10.1007/978-3-030-88942-5</a>.'
  mla: Damke, Clemens, and Eyke Hüllermeier. “Ranking Structured Objects with Graph
    Neural Networks.” <i>Proceedings of The 24th International Conference on Discovery
    Science (DS 2021)</i>, edited by Carlos Soares and Luis Torgo, vol. 12986, Springer,
    2021, pp. 166–80, doi:<a href="https://doi.org/10.1007/978-3-030-88942-5">10.1007/978-3-030-88942-5</a>.
  short: 'C. Damke, E. Hüllermeier, in: C. Soares, L. Torgo (Eds.), Proceedings of
    The 24th International Conference on Discovery Science (DS 2021), Springer, 2021,
    pp. 166–180.'
conference:
  end_date: 2021-10-13
  location: Halifax, Canada
  name: 24th International Conference on Discovery Science
  start_date: 2021-10-11
date_created: 2021-11-11T14:15:18Z
date_updated: 2022-04-11T22:08:12Z
department:
- _id: '355'
doi: 10.1007/978-3-030-88942-5
editor:
- first_name: Carlos
  full_name: Soares, Carlos
  last_name: Soares
- first_name: Luis
  full_name: Torgo, Luis
  last_name: Torgo
external_id:
  arxiv:
  - '2104.08869'
intvolume: '     12986'
keyword:
- Graph-structured data
- Graph neural networks
- Preference learning
- Learning to rank
language:
- iso: eng
page: 166-180
publication: Proceedings of The 24th International Conference on Discovery Science
  (DS 2021)
publication_identifier:
  isbn:
  - '9783030889418'
  - '9783030889425'
  issn:
  - 0302-9743
  - 1611-3349
publication_status: published
publisher: Springer
quality_controlled: '1'
series_title: Lecture Notes in Computer Science
status: public
title: Ranking Structured Objects with Graph Neural Networks
type: conference
user_id: '48192'
volume: 12986
year: '2021'
...
---
_id: '19953'
abstract:
- lang: eng
  text: Current GNN architectures use a vertex neighborhood aggregation scheme, which
    limits their discriminative power to that of the 1-dimensional Weisfeiler-Lehman
    (WL) graph isomorphism test. Here, we propose a novel graph convolution operator
    that is based on the 2-dimensional WL test. We formally show that the resulting
    2-WL-GNN architecture is more discriminative than existing GNN approaches. This
    theoretical result is complemented by experimental studies using synthetic and
    real data. On multiple common graph classification benchmarks, we demonstrate
    that the proposed model is competitive with state-of-the-art graph kernels and
    GNNs.
author:
- first_name: Clemens
  full_name: Damke, Clemens
  id: '48192'
  last_name: Damke
  orcid: 0000-0002-0455-0048
- first_name: Vitaly
  full_name: Melnikov, Vitaly
  id: '58747'
  last_name: Melnikov
- first_name: Eyke
  full_name: Hüllermeier, Eyke
  id: '48129'
  last_name: Hüllermeier
citation:
  ama: 'Damke C, Melnikov V, Hüllermeier E. A Novel Higher-order Weisfeiler-Lehman
    Graph Convolution. In: Jialin Pan S, Sugiyama M, eds. <i>Proceedings of the 12th
    Asian Conference on Machine Learning (ACML 2020)</i>. Vol 129. Proceedings of
    Machine Learning Research. Bangkok, Thailand: PMLR; 2020:49-64.'
  apa: 'Damke, C., Melnikov, V., &#38; Hüllermeier, E. (2020). A Novel Higher-order
    Weisfeiler-Lehman Graph Convolution. In S. Jialin Pan &#38; M. Sugiyama (Eds.),
    <i>Proceedings of the 12th Asian Conference on Machine Learning (ACML 2020)</i>
    (Vol. 129, pp. 49–64). Bangkok, Thailand: PMLR.'
  bibtex: '@inproceedings{Damke_Melnikov_Hüllermeier_2020, place={Bangkok, Thailand},
    series={Proceedings of Machine Learning Research}, title={A Novel Higher-order
    Weisfeiler-Lehman Graph Convolution}, volume={129}, booktitle={Proceedings of
    the 12th Asian Conference on Machine Learning (ACML 2020)}, publisher={PMLR},
    author={Damke, Clemens and Melnikov, Vitaly and Hüllermeier, Eyke}, editor={Jialin
    Pan, Sinno and Sugiyama, MasashiEditors}, year={2020}, pages={49–64}, collection={Proceedings
    of Machine Learning Research} }'
  chicago: 'Damke, Clemens, Vitaly Melnikov, and Eyke Hüllermeier. “A Novel Higher-Order
    Weisfeiler-Lehman Graph Convolution.” In <i>Proceedings of the 12th Asian Conference
    on Machine Learning (ACML 2020)</i>, edited by Sinno Jialin Pan and Masashi Sugiyama,
    129:49–64. Proceedings of Machine Learning Research. Bangkok, Thailand: PMLR,
    2020.'
  ieee: C. Damke, V. Melnikov, and E. Hüllermeier, “A Novel Higher-order Weisfeiler-Lehman
    Graph Convolution,” in <i>Proceedings of the 12th Asian Conference on Machine
    Learning (ACML 2020)</i>, Bangkok, Thailand, 2020, vol. 129, pp. 49–64.
  mla: Damke, Clemens, et al. “A Novel Higher-Order Weisfeiler-Lehman Graph Convolution.”
    <i>Proceedings of the 12th Asian Conference on Machine Learning (ACML 2020)</i>,
    edited by Sinno Jialin Pan and Masashi Sugiyama, vol. 129, PMLR, 2020, pp. 49–64.
  short: 'C. Damke, V. Melnikov, E. Hüllermeier, in: S. Jialin Pan, M. Sugiyama (Eds.),
    Proceedings of the 12th Asian Conference on Machine Learning (ACML 2020), PMLR,
    Bangkok, Thailand, 2020, pp. 49–64.'
conference:
  end_date: 2020-11-20
  location: Bangkok, Thailand
  name: Asian Conference on Machine Learning
  start_date: 2020-11-18
date_created: 2020-10-08T10:48:38Z
date_updated: 2022-01-06T06:54:17Z
ddc:
- '006'
department:
- _id: '355'
editor:
- first_name: Sinno
  full_name: Jialin Pan, Sinno
  last_name: Jialin Pan
- first_name: Masashi
  full_name: Sugiyama, Masashi
  last_name: Sugiyama
external_id:
  arxiv:
  - '2007.00346'
file:
- access_level: open_access
  content_type: application/pdf
  creator: cdamke
  date_created: 2020-10-08T10:54:48Z
  date_updated: 2020-10-08T11:21:00Z
  file_id: '19954'
  file_name: damke20.pdf
  file_size: 771137
  relation: main_file
- access_level: open_access
  content_type: application/pdf
  creator: cdamke
  date_created: 2020-10-08T10:54:59Z
  date_updated: 2020-10-08T11:24:29Z
  file_id: '19955'
  file_name: damke20-supp.pdf
  file_size: 613163
  relation: supplementary_material
file_date_updated: 2020-10-08T11:24:29Z
has_accepted_license: '1'
intvolume: '       129'
keyword:
- graph neural networks
- Weisfeiler-Lehman test
- cycle detection
language:
- iso: eng
oa: '1'
page: 49-64
place: Bangkok, Thailand
publication: Proceedings of the 12th Asian Conference on Machine Learning (ACML 2020)
publication_status: published
publisher: PMLR
quality_controlled: '1'
series_title: Proceedings of Machine Learning Research
status: public
title: A Novel Higher-order Weisfeiler-Lehman Graph Convolution
type: conference
user_id: '48192'
volume: 129
year: '2020'
...