Neuro-Symbolic Spatio-Temporal Reasoning

  title={Neuro-Symbolic Spatio-Temporal Reasoning},
  author={Jae Hee Lee and Michael Sioutis and Kyra Ahrens and Marjan Alirezaie and Matthias Kerzel and Stefan Wermter},
Knowledge about space and time is an essential prerequisite for solving problems in the physical world: An artificial intelligence (AI) agent, when situated in a physical environment and interacting with objects, often needs to reason about positions of and relations between objects; and as soon as the agent plans its actions to solve a task, it needs to consider the temporal aspect (e.g., what actions to perform over time). Spatio-temporal knowledge, however, is required beyond interacting… 

Figures and Tables from this paper



A Spatial Logic based on Regions and Connection

An interval logic for reasoning about space is described, which supports a simpler ontology, has fewer functions and relations, yet does not su er in terms of its useful expressiveness.

A Generalised Approach for Encoding and Reasoning with Qualitative Theories in Answer Set Programming

Experimental results show that the proposed encodings outperform one of the two reasoners, but fall behind the other, an acceptable trade-off given the added benefits of handling any type of reasoning as well as the interpretability of logic programs.

A A Survey of Qualitative Spatial and Temporal Calculi — Algebraic and Computational Properties

  • 2016

Non-monotonic spatial reasoning with answer set programming modulo theories* †

Answer Set Programming Modulo Theories (ASPMT)(QS), a novel approach and fully implemented prototype for non-monotonic spatial reasoning — a crucial requirement within dynamic spatial systems — based on ASPMT is presented.

Compositional Attention Networks for Machine Reasoning

The MAC network is presented, a novel fully differentiable neural network architecture, designed to facilitate explicit and expressive reasoning that is computationally-efficient and data-efficient, in particular requiring 5x less data than existing models to achieve strong results.

Allen's Interval Algebra Makes the Difference

This paper presents a novel encoding of Interval Algebra using answer-set programming (ASP) extended by difference constraints, i.e., the fragment abbreviated as ASP(DL), and suggests that analogous encodings can be devised for other point-based calculi, too.

Semantic Referee: A Neural-Symbolic Framework for Enhancing Geospatial Semantic Segmentation

A semantic referee is proposed, which is able to extract qualitative features of the errors emerging from deep machine learning frameworks and suggest corrections and how to improve the performance of semantic segmentation for satellite imagery data.

PTR: A Benchmark for Part-based Conceptual, Relational, and Physical Reasoning

A new large-scale diagnostic visual reasoning dataset named PTR is introduced, which contains around 70k RGBD synthetic images with ground truth object and part level annotations regarding semantic instance segmentation, color attributes, spatial and geometric relationships, and certain physical properties such as stability.

Mereotopological Correction of Segmentation Errors in Histological Imaging

In this paper we describe mereotopological methods to programmatically correct image segmentation errors, in particular those that fail to fulfil expected spatial relations in digitised histological

A Model-Theoretic View on Qualitative Constraint Reasoning

A model-theoretic perspective on qualitative constraint reasoning is presented and the significance of omega-categoricity for qualitative reasoning, of primitive positive interpretations for complexity analysis, and of Datalog as a unifying language for describing local consistency algorithms are discussed.