Conflict-Aware Pseudo Labeling via Optimal Transport for Entity
Alignment
- URL: http://arxiv.org/abs/2209.01847v1
- Date: Mon, 5 Sep 2022 09:14:01 GMT
- Title: Conflict-Aware Pseudo Labeling via Optimal Transport for Entity
Alignment
- Authors: Qijie Ding, Daokun Zhang, Jie Yin
- Abstract summary: We propose a Conflict-aware Pseudo Labeling via Optimal Transport model (CPL-OT) for entity alignment.
CPL-OT is composed of two key components-entity embedding learning with global-local aggregation and iterative conflict-aware pseudo labeling.
It can markedly outperform state-of-the-art baselines under both settings with and without prior alignment seeds.
- Score: 4.5591913587473964
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Entity alignment aims to discover unique equivalent entity pairs with the
same meaning across different knowledge graphs (KG). It has been a compelling
but challenging task for knowledge integration or fusion. Existing models have
primarily focused on projecting KGs into a latent embedding space to capture
inherent semantics between entities for entity alignment. However, the adverse
impacts of alignment conflicts have been largely overlooked during training,
thus limiting the entity alignment performance. To address this issue, we
propose a novel Conflict-aware Pseudo Labeling via Optimal Transport model
(CPL-OT) for entity alignment. The key idea of CPL-OT is to iteratively
pseudo-label alignment pairs empowered with conflict-aware Optimal Transport
modeling to boost the precision of entity alignment. CPL-OT is composed of two
key components-entity embedding learning with global-local aggregation and
iterative conflict-aware pseudo labeling-that mutually reinforce each other. To
mitigate alignment conflicts during pseudo labeling, we propose to use optimal
transport (OT) as an effective means to warrant one-to-one entity alignment
between two KGs with the minimal overall transport cost. The transport cost is
calculated as the rectified distance between entity embeddings obtained via
graph convolution augmented with global-level semantics. Extensive experiments
on benchmark datasets show that CPL-OT can markedly outperform state-of-the-art
baselines under both settings with and without prior alignment seeds.
Related papers
- Entity Alignment with Unlabeled Dangling Cases [49.86384156476041]
We propose a novel GNN-based dangling detection and entity alignment framework.
While the two tasks share the same GNN, the detected dangling entities are removed in the alignment.
Our framework is featured by a designed entity and relation attention mechanism for selective neighborhood aggregation in representation learning.
arXiv Detail & Related papers (2024-03-16T17:21:58Z) - Unsupervised Cross-Domain Image Retrieval via Prototypical Optimal Transport [24.141959555741696]
Unsupervised cross-domain image retrieval aims to retrieve images sharing the same category across diverse domains without relying on labeled data.
Prior approaches have typically decomposed the UCIR problem into two distinct tasks: intra-domain representation learning and cross-domain feature alignment.
This paper introduces ProtoOT, a novel Optimal Transport formulation explicitly tailored for UCIR.
arXiv Detail & Related papers (2024-02-28T15:31:45Z) - Semi-DETR: Semi-Supervised Object Detection with Detection Transformers [105.45018934087076]
We analyze the DETR-based framework on semi-supervised object detection (SSOD)
We present Semi-DETR, the first transformer-based end-to-end semi-supervised object detector.
Our method outperforms all state-of-the-art methods by clear margins.
arXiv Detail & Related papers (2023-07-16T16:32:14Z) - Combating Confirmation Bias: A Unified Pseudo-Labeling Framework for
Entity Alignment [29.18600512063287]
We propose a Unified Pseudo-Labeling framework for Entity Alignment (UPL-EA)
UPL-EA explicitly eliminates pseudo-labeling errors to boost the accuracy of entity alignment.
The effectiveness of UPL-EA in eliminating pseudo-labeling errors is both theoretically supported and experimentally validated.
arXiv Detail & Related papers (2023-07-05T07:32:34Z) - From Alignment to Entailment: A Unified Textual Entailment Framework for
Entity Alignment [17.70562397382911]
Existing methods usually encode the triples of entities as embeddings and learn to align the embeddings.
We transform both triples into unified textual sequences, and model the EA task as a bi-directional textual entailment task.
Our approach captures the unified correlation pattern of two kinds of information between entities, and explicitly models the fine-grained interaction between original entity information.
arXiv Detail & Related papers (2023-05-19T08:06:50Z) - Optimizing Bi-Encoder for Named Entity Recognition via Contrastive
Learning [80.36076044023581]
We present an efficient bi-encoder framework for named entity recognition (NER)
We frame NER as a metric learning problem that maximizes the similarity between the vector representations of an entity mention and its type.
A major challenge to this bi-encoder formulation for NER lies in separating non-entity spans from entity mentions.
arXiv Detail & Related papers (2022-08-30T23:19:04Z) - Disentangled Federated Learning for Tackling Attributes Skew via
Invariant Aggregation and Diversity Transferring [104.19414150171472]
Attributes skews the current federated learning (FL) frameworks from consistent optimization directions among the clients.
We propose disentangled federated learning (DFL) to disentangle the domain-specific and cross-invariant attributes into two complementary branches.
Experiments verify that DFL facilitates FL with higher performance, better interpretability, and faster convergence rate, compared with SOTA FL methods.
arXiv Detail & Related papers (2022-06-14T13:12:12Z) - Semi-constraint Optimal Transport for Entity Alignment with Dangling
Cases [6.755145435406154]
We propose an unsupervised method called Semi-constraint Optimal Transport for Entity Alignment in Dangling cases (SoTead)
Our main idea is to model the entity alignment between two KGs as an optimal transport problem from one KG's entities to the others.
In the experimental part, we first show the superiority of SoTead on a commonly-used entity alignment dataset.
arXiv Detail & Related papers (2022-03-11T04:20:18Z) - Semi-supervised Domain Adaptive Structure Learning [72.01544419893628]
Semi-supervised domain adaptation (SSDA) is a challenging problem requiring methods to overcome both 1) overfitting towards poorly annotated data and 2) distribution shift across domains.
We introduce an adaptive structure learning method to regularize the cooperation of SSL and DA.
arXiv Detail & Related papers (2021-12-12T06:11:16Z) - Simultaneous Semantic Alignment Network for Heterogeneous Domain
Adaptation [67.37606333193357]
We propose aSimultaneous Semantic Alignment Network (SSAN) to simultaneously exploit correlations among categories and align the centroids for each category across domains.
By leveraging target pseudo-labels, a robust triplet-centroid alignment mechanism is explicitly applied to align feature representations for each category.
Experiments on various HDA tasks across text-to-image, image-to-image and text-to-text successfully validate the superiority of our SSAN against state-of-the-art HDA methods.
arXiv Detail & Related papers (2020-08-04T16:20:37Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.