A shallow neural model for relation prediction

Demir, Caglar; Moussallem, Diego; Ngomo, Axel-Cyrille Ngonga

doi:10.1109/icsc50631.2021.00038

Cited by 8 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To train Drill, we first generate 10 learning problems for each benchmark datasets described in Section 5.1. During our experiments, we used the pretrained embeddings of input knowledge graphs provided by [10,9]. We used ConEx embeddings of Family, Biopax, Mutagenesis and Shallom embeddings of Carcinogenesis.…”

Section: Methodsmentioning

confidence: 99%

“…We used ConEx embeddings of Family, Biopax, Mutagenesis and Shallom embeddings of Carcinogenesis. For more details about configurations of models, we refer project pages of [10,9]. For each generated learning problem, we trained Drill in an ǫgreedy fashion by using the the following configuration: ADAM optimizer with learning rate of .01, mini-batches of size 512, number of episodes set to 100, an epsilon decay of .01, a discounting factor γ of .99, 32 input channels, and (3 × 3)…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

DRILL-- Deep Reinforcement Learning for Refinement Operators in $\mathcal{ALC}$

Demir¹,

Ngomo²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Approaches based on refinement operators have been successfully applied to class expression learning on RDF knowledge graphs. These approaches often need to explore a large number of concepts to find adequate hypotheses. This need arguably stems from current approaches relying on myopic heuristic functions to guide their search through an infinite concept space. In turn, deep reinforcement learning provides effective means to address myopia by estimating how much discounted cumulated future reward states promise. In this work, we leverage deep reinforcement learning to accelerate the learning of concepts in ALC by proposing Drill-a novel class expression learning approach that uses a convolutional deep Q-learning model to steer its search. By virtue of its architecture, Drill is able to compute the expected discounted cumulated future reward of more than 10 3 class expressions in a second on standard hardware. We evaluate Drill on four benchmark datasets against state-of-the-art approaches. Our results suggest that Drill converges to goal states at least 2.7× faster than state-of-the-art models on all benchmark datasets. We provide an open-source implementation of our approach, including training and evaluation scripts as well as pretrained models. 1

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

DRILL-- Deep Reinforcement Learning for Refinement Operators in $\mathcal{ALC}$

Demir¹,

Ngomo²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Scholarly Publications: Our framework have been effectively used to learn knowledge graphs embeddings in several published works [3,4,[13][14][15][16], and [17].…”

Section: Software Impactmentioning

confidence: 99%

Hardware-agnostic computation for large-scale knowledge graph embeddings

Demir

Ngomo²

2022

Software Impacts

Self Cite

View full text Add to dashboard Cite

“…Scholarly Publications: Our framework have been effectively used to learn knowledge graphs embeddings in several published works [13], [3], [14], [4], [15], [16], and [17].…”

Section: Software Impactmentioning

confidence: 99%

Hardware-agnostic Computation for Large-scale Knowledge Graph Embeddings

Demir¹,

Ngomo²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Knowledge graph embedding research has mainly focused on learning continuous representations of knowledge graphs towards the link prediction problem. Recently developed frameworks can be effectively applied in research related applications. Yet, these frameworks do not fulfill many requirements of real-world applications. As the size of the knowledge graph grows, moving computation from a commodity computer to a cluster of computers in these frameworks becomes more challenging. Finding suitable hyperparameter settings w.r.t. time and computational budgets are left to practitioners. In addition, the continual learning aspect in knowledge graph embedding frameworks is often ignored, although continual learning plays an important role in many real-world (deep) learning-driven applications. Arguably, these limitations explain the lack of publicly available knowledge graph embedding models for large knowledge graphs. We developed a framework based on the frameworks DASK, Pytorch Lightning and Hugging Face to compute embeddings for large-scale knowledge graphs in a hardware-agnostic manner, which is able to address real-world challenges pertaining to the scale of real application. We provide an open-source version of our framework along with a hub of pre-trained models having more than 11.4 B parameters 2 .

show abstract

A shallow neural model for relation prediction

Cited by 8 publications

References 28 publications

DRILL-- Deep Reinforcement Learning for Refinement Operators in $\mathcal{ALC}$

DRILL-- Deep Reinforcement Learning for Refinement Operators in $\mathcal{ALC}$

Hardware-agnostic computation for large-scale knowledge graph embeddings

Hardware-agnostic Computation for Large-scale Knowledge Graph Embeddings

Contact Info

Product

Resources

About