S
Seyed Mehran Kazemi
Researcher at University of British Columbia
Publications - 41
Citations - 1541
Seyed Mehran Kazemi is an academic researcher from University of British Columbia. The author has contributed to research in topics: Probabilistic logic & Inference. The author has an hindex of 14, co-authored 35 publications receiving 894 citations. Previous affiliations of Seyed Mehran Kazemi include Google.
Papers
More filters
Proceedings Article
SimplE embedding for link prediction in knowledge graphs
Seyed Mehran Kazemi,David Poole +1 more
TL;DR: It is proved SimplE is fully expressive and derive a bound on the size of its embeddings for full expressivity and shown empirically that, despite its simplicity, SimplE outperforms several state-of-the-art tensor factorization techniques.
Journal Article
Representation Learning for Dynamic Graphs: A Survey
Seyed Mehran Kazemi,Rishab Goel,Kshitij Jain,Ivan Kobyzev,Akshay Sethi,Peter Forsyth,Pascal Poupart +6 more
TL;DR: This survey describes existing models from an encoder-decoder perspective, categorize these encoders and decoders based on the techniques they employ, and analyzes the approaches in each category.
Journal ArticleDOI
Diachronic Embedding for Temporal Knowledge Graph Completion
TL;DR: Novel models for temporal KG completion are built through equipping static models with a diachronic entity embedding function which provides the characteristics of entities at any point in time where only static entity features are provided.
Posted Content
Time2Vec: Learning a Vector Representation of Time
Seyed Mehran Kazemi,Rishab Goel,Sepehr Eghbali,Janahan Ramanan,Jaspreet Sahota,Sanjay Thakur,Stella Wu,Cathal Smyth,Pascal Poupart,Marcus A. Brubaker +9 more
TL;DR: This paper provides a model-agnostic vector representation for time, called Time2Vec, that can be easily imported into many existing and future architectures and improve their performances.
Posted Content
SimplE Embedding for Link Prediction in Knowledge Graphs
Seyed Mehran Kazemi,David Poole +1 more
TL;DR: SimplE as mentioned in this paper is an extension of the Canonical polyadic decomposition to allow the two embeddings of each entity to be learned dependently, and the complexity of SimplE grows linearly with the size of embedding.