Search Results for author: Ralf Möller

Found 10 papers, 0 papers with code

Improved Convergence Speed of Fully Symmetric Learning Rules for Principal Component Analysis

no code implementations18 Jul 2020 Ralf Möller

Fully symmetric learning rules for principal component analysis can be derived from a novel objective function suggested in our previous work.

Derivation of Symmetric PCA Learning Rules from a Novel Objective Function

no code implementations24 May 2020 Ralf Möller

However, for a subspace with multiple axes, the optimization leads to PSA learning rules which only converge to axes spanning the principal subspace but not to the principal eigenvectors.

Derivation of Coupled PCA and SVD Learning Rules from a Newton Zero-Finding Framework

no code implementations25 Mar 2020 Ralf Möller

A method to derive coupled learning rules from information criteria by Newton optimization is known.

Exploring Unknown Universes in Probabilistic Relational Models

no code implementations7 Jan 2020 Tanya Braun, Ralf Möller

Large probabilistic models are often shaped by a pool of known individuals (a universe) and relations between them.

Taming Reasoning in Temporal Probabilistic Relational Models

no code implementations16 Nov 2019 Marcel Gehrke, Ralf Möller, Tanya Braun

Evidence often grounds temporal probabilistic relational models over time, which makes reasoning infeasible.

Preventing Unnecessary Groundings in the Lifted Dynamic Junction Tree Algorithm

no code implementations2 Jul 2018 Marcel Gehrke, Tanya Braun, Ralf Möller

The lifted dynamic junction tree algorithm (LDJT) efficiently answers filtering and prediction queries for probabilistic relational temporal models by building and then reusing a first-order cluster representation of a knowledge base for multiple queries and time steps.

Fusing First-order Knowledge Compilation and the Lifted Junction Tree Algorithm

no code implementations2 Jul 2018 Tanya Braun, Ralf Möller

Standard approaches for inference in probabilistic formalisms with first-order constructs include lifted variable elimination (LVE) for single queries as well as first-order knowledge compilation (FOKC) based on weighted model counting.

Answering Hindsight Queries with Lifted Dynamic Junction Trees

no code implementations2 Jul 2018 Marcel Gehrke, Tanya Braun, Ralf Möller

The lifted dynamic junction tree algorithm (LDJT) efficiently answers filtering and prediction queries for probabilistic relational temporal models by building and then reusing a first-order cluster representation of a knowledge base for multiple queries and time steps.

Towards Analytics Aware Ontology Based Access to Static and Streaming Data (Extended Version)

no code implementations18 Jul 2016 Evgeny Kharlamov, Yannis Kotidis, Theofilos Mailis, Christian Neuenstadt, Charalampos Nikolaou, Özgür Özcep, Christoforos Svingos, Dmitriy Zheleznyakov, Sebastian Brandt, Ian Horrocks, Yannis Ioannidis, Steffen Lamparter, Ralf Möller

Real-time analytics that requires integration and aggregation of heterogeneous and distributed streaming and static data is a typical task in many industrial scenarios such as diagnostics of turbines in Siemens.

Cannot find the paper you are looking for? You can Submit a new open access paper.