\( \newcommand{\R}{\mathbb{R}} \newcommand{\ten}[1]{\mathbf{#1}} \newcommand{\tenp}{\otimes} \newcommand{\bigtenp}{\bigotimes} \newcommand{\Set}[1]{\mathcal{#1}} \newcommand{\re}[3]{\vec{a}^T_{#1}R^{\vphantom{T}}_{#3}\vec{a}^{\vphantom{T}}_{#2}} \newcommand{\ttm}{\times} \newcommand{\vect}{\mathbf{vec}} \newcommand{\pred}[1]{\widehat{#1}} \newcommand{\BigO}{\mathcal{O}} \newcommand{\hadamard}{*} \newcommand{\discrep}{\mathcal{D}} \newcommand{\unfold}[2]{#1_{(#2)}} \newcommand{\from}{\sim} \renewcommand{\vec}[1]{\mathbf{#1}} \)

Maximilian Nickel


Bio I am currently a postdoctoral fellow with the Laboratory for Computational and Statistical Learning (LCSL), hosted at the Center for Biological and Computational Learning (CBCL), Massachusetts Institute of Technology (MIT). In 2013, I received my Ph.D. summa cum laude from the Department for Informatics of the Ludwig-Maximilians-University of Munich under supervision of Volker Tresp. I received a diploma degree with honors in computer science from the Ludwig Maximilian University Munich in 2009.

Research Interests

Scroll down for a full list of publications and code.


Publications

A list of selected publications in peer-reviewed conferences and journals, as well as book chapter, tutorials, and workshops.

You can also find my publications on Google Scholar

Tutorials

    • Maximilian Nickel and Volker Tresp
    • Machine Learning on Linked Data: Tensors and their Applications in Graph-Structured Domains
    • Tutorial at ISWC 2012

Conference Papers

    • , ,
    • Holographic Embeddings Knowledge Graphs
    • Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16)
      • , , ,
      • A Review of Relational Machine Learning for Knowledge Graphs
      • Proceedings of the IEEE, 2016.
      • , ,
      • Learning from Latent and Observable Patterns in Multi-Relational Data
      • Advances in Neural Information Processing Systems (NIPS), 2014.
      • , ,
      • Querying Factorized Probabilistic Triple Databases
      • Proceedings of the 13th International Semantic Web Conferende (ISWC), 2014.
      • ,
      • Tensor Factorization for Multi-Relational Learning
      • Machine Learning and Knowledge Discovery in Databases, Springer, 2013.
      • ,
      • An Analysis of Tensor Models for Learning from Structured Data.
      • Machine Learning and Knowledge Discovery in Databases, Springer, 2013.
      • ,
      • Logistic Tensor Factorization for Multi-Relational Data
      • in ICML 2013 Workshop - Structured Learning: Inferring Graphs from Structured and Unstructured Inputs (SLG2013), Atlanta, USA.
      • , ,
      • Factorizing YAGO: Scalable Machine Learning for Linked Data
      • in Proceedings of the 21st International World Wide Web Conference (WWW), 2012, Lyon, France
      • , , ,
      • Combining Information Extraction, Deductive Reasoning and Machine Learning for Relation Prediction
      • in Proceedings of the 9th Extended Semantic Web Conference (ESWC), 2012
      • , , , ,
      • Scalable Relation Prediction Exploiting Both Intrarelational Correlation and Contextual Information
      • Proceedings of the ECML/PKDD, 2012
      • , ,
      • A Three-Way Model for Collective Learning on Multi-Relational Data
      • in Proceedings of the 28th International Conference on Machine Learning (ICML), 2011 , Bellevue, WA, USA.
      • ,
      • Learning Taxonomies from Multi-Relational Data via Hierarchical Link-Based Clustering
      • in NIPS 2011 Workshop - Learning Semantics, 2011 , Granada, Spain
      • , , ,
      • A Scalable Kernel Approach to Learning in Semantic Graphs with Applications to Linked Data
      • in Proceedings of the 1st Workshop on Mining the Future Internet, 2010.
      • ,
      • Three-Way DEDICOM for Relational Learning
      • in NIPS 2010 Workshop - Tensors, Kernels and Machine Learning, 2010, Whistler, Canada.

Journal Papers

    • Yi Huang, Volker Tresp, Maximilian Nickel, Achim Rettinger, and Hans-Peter Kriegel
    • A Scalable Approach for Statistical Learning in Semantic Graphs
    • Semantic Web – Interoperability, Usability, Applicability (SWJ), 2013

Book Chapters

    • Volker Tresp, Yi Huang, and Maximilian Nickel
    • Querying the Web with Statistical Machine Learning
    • To be published as a chapter in a book reviewing the results of the THESEUS project, 2013

Thesis

    • Maximilian Nickel
    • Tensor Factorization for Relational Learning
    • Ludwig Maximilian University, Munich, 2013

Code & Projects

RESCAL

RESCAL is a tensor factorization for large-scale relational learning from Linked Data, multi-relational data and large multigraphs. RESCAL offers state-of-the-art relational learning results combined with high scalability, such that it can be applied to data consisting of millions of entities, hundreds of relations, and billions of known facts.

Factorization Model
Let $\ten{X}$ be an adjacency tensor representing a multigraph. RESCAL factorizes $\ten{X}$ into \[ \ten{X} \approx \ten{R} \ttm_1 A \ttm_2 A, \] or equivalently in elementwise notation into \[ x_{ijk} \approx \re{i}{j}{k}. \]
RESCAL factorization model

scikit-tensor

scikit-tensor is a Python library for multilinear algebra and tensor factorizations. It includes routines to compute factorizations such as the Tucker decomposition, CP, RESCAL and others. The focus of the library lies on easy-to-use code for fast prototyping as well as high performance and scalability.