Home Organizers Goals Content StarAI

Tutorial on StarAI

Stochastic Relational AI at KI-2018

The 41st German Conference on Artificial Intelligence
September 24-28th, 2018, Berlin, Germany

Organizers

The tuturial is organized by Tanya Braun, Kristian Kersting, and Ralf Möller

Tanya Braun is a research associate at Universität zu Lübeck,
Kristian Kersting is a professor for Machine Learning at Darmstadt University, and
Ralf Möller is a professor for Information Systems at Universität zu Lübeck

Goals

The goal of the tutorial is to give an overview about recent developments in probabilistic relational modeling and reasoning

Probabilistic relational modeling for artificial intelligence (also called stochastic relational AI) has attracted the interest of a huge number of researchers in the last ten years. Assuming a background in probability theory as well as in basics about Bayesian or Markovian networks, we shortly introduce specific probabilistic relational modeling approaches and focus on exact and approximative inference algorithms. Recent developments in the area of learning techniques will be presented as well.

Contents

We cover StarAI from foundations to applications

  1. Introduction (Kristian)
    • StarAI
    • SystemsAI
  2. Probabilistic relational modeling (Ralf)
    • Semantics (grounded-distributional, density-based, maximum entropy based)
    • Inference problems and applications
    • Algorithms and systems
    • Scalability
      • Lmited expressivity: probabilistic databases
      • Knowledge compilation: linear programming and weighted model counting
      • Approximation: belief propagation, TensorLog
  3. Scalability by lifting
    • Exact lifted inference (Tanya)
      • Lifted Junction Tree Algorithm
      • Lifted Dynamic Juntion Tree Algorithm
    • Approximate lifted inference (Kristian)
  4. Learning (Kristian)
    • Parameter learning (stochastic gradient descent)
    • Structure learning
    • Relational reinforcement learning
  5. Summary (Kristian)

StarAI Lays the Foundation for Artificial Intelligence