Biography

Afshin Sadeghi

Dr. Afshin Sadeghi is a PhD in the field of machine learning. In the years 2016-2018, he worked in University of Bonn as a researcher in Semantic Web and Knowledge Graphs. He joined Fraunhofer Institute in 2018 as a researcher with focus on modeling Knowledge Graphs. His research centers around representation learning methods for knowledge graphs and their applications in NLP and Industrial Graphs.

Selected Publications

Embedding Knowledge Graphs Attentive to Positional and Centrality Qualities (ECML 2021)

This work proposes a novel KGE method named Graph Feature Attentive Neural Network (GFA-NN) that computes graphical features of entities. As a consequence, the resulting embeddings are attentive to two types of global network features.

Read the Complete Paper Source Code

MDE: Multiple Distance Embeddings for Link Prediction in Knowledge Graphs (ECAI 2020)

MDE proposes Multi objective learning for KG embedding. By learning independent embedding vectors for each of the terms we can collectively train and predict using contradicting distance terms. We further demonstrate that MDE allows modeling relations with (anti)symmetry, inversion, and composition patterns. We pro- pose MDE as a neural network model that allows us to map non- linear relations between the embedding vectors and the expected out- put of the score function.

Read the Complete Paper Source Code

(NeurIPS 2021)

Relational Pattern Benchmarking on the Knowledge Graph Link Prediction Task

Despite all the advancements in KGs, they plummet when it comes to completeness. Link Prediction based on KG embeddings targets the sparsity and incompleteness of KGs. Available datasets for Link Prediction do not consider different graph patterns, making it difficult to measure the performance of link prediction models on different KG settings. This paper presents a diverse set of pragmatic datasets to facilitate flexible and problem-tailored Link Prediction and Knowledge Graph Embeddings research. We define graph relational patterns, from being entirely inductive in one set to being transductive in the other.

Read the Complete Paper Source Code