Improving Reinforcement Learning Results with Qualitative Spatial Representation

2017 ◽  
Author(s):  
Thiago Pedro Donadon Homem ◽  
Danilo Hernani Perico ◽  
Paulo Eduardo Santos ◽  
Anna Helena Reali Costa ◽  
Reinaldo Augusto da Costa Bianchi
Keyword(s):  
Reinforcement Learning ◽  
Spatial Representation ◽  
Qualitative Spatial Representation

scholarly journals LEARNING TO BEHAVE IN SPACE: A QUALITATIVE SPATIAL REPRESENTATION FOR ROBOT NAVIGATION WITH REINFORCEMENT LEARNING

2008 ◽  
Vol 17 (03) ◽  
pp. 465-482 ◽  
Author(s):  
LUTZ FROMMBERGER
Keyword(s):  
Reinforcement Learning ◽  
Spatial Representation ◽  
Robot Navigation ◽  
Structural Knowledge ◽  
Navigation Strategy ◽  
The World ◽  
Representation Of Space ◽  
Moving Direction ◽  
Qualitative Spatial Representation ◽  
Circular Order

The representation of the surrounding world plays an important role in robot navigation, especially when reinforcement learning is applied. This work uses a qualitative abstraction mechanism to create a representation of space consisting of the circular order of detected landmarks and the relative position of walls towards the agent's moving direction. The use of this representation does not only empower the agent to learn a certain goal-directed navigation strategy faster compared to metrical representations, but also facilitates reusing structural knowledge of the world at different locations within the same environment. Acquired policies are also applicable in scenarios with different metrics and corridor angles. Furthermore, gained structural knowledge can be separated, leading to a generally sensible navigation behavior that can be transferred to environments lacking landmark information and/or totally unknown environments.

Neuronal Activity in the Rodent Dorsal Striatum in Sequential Navigation: Separation of Spatial and Reward Responses on the Multiple T Task

2004 ◽  
Vol 91 (5) ◽  
pp. 2259-2272 ◽  
Author(s):  
Neil Schmitzer-Torbert ◽  
A. David Redish
Keyword(s):  
Reinforcement Learning ◽  
Spatial Representation ◽  
Sensory Input ◽  
Learning Algorithm ◽  
Dorsal Striatum ◽  
Food Delivery ◽  
Simple Motor ◽  
Experimental Approaches ◽  
Highly Correlated ◽  
To Receive

The striatum plays an important role in “habitual” learning and memory and has been hypothesized to implement a reinforcement-learning algorithm to select actions to perform given the current sensory input. Many experimental approaches to striatal activity have made use of temporally structured tasks, which imply that the striatal representation is temporal. To test this assumption, we recorded neurons in the dorsal striatum of rats running a sequential navigation task: the multiple T maze. Rats navigated a sequence of four T maze turns to receive food rewards delivered in two locations. The responses of neurons that fired phasically were examined. Task-responsive phasic neurons were active as rats ran on the maze (maze-responsive) or during reward receipt (reward-responsive). Neither mazenor reward-responsive neurons encoded simple motor commands: maze-responses were not well correlated with the shape of the rat's path and most reward-responsive neurons did not fire at similar rates at both food-delivery sites. Maze-responsive neurons were active at one or more locations on the maze, but these responses did not cluster at spatial landmarks such as turns. Across sessions the activity of maze-responsive neurons was highly correlated when rats ran the same maze. Maze-responses encoded the location of the rat on the maze and imply a spatial representation in the striatum in a task with prominent spatial demands. Maze-responsive and reward-responsive neurons were two separate populations, suggesting a divergence in striatal information processing of navigation and reward.

scholarly journals A survey of qualitative spatial representations

2013 ◽  
Vol 30 (1) ◽  
pp. 106-136 ◽  
Author(s):  
Juan Chen ◽  
Anthony G. Cohn ◽  
Dayou Liu ◽  
Shengsheng Wang ◽  
Jihong Ouyang ◽  
...  
Keyword(s):  
Spatial Representation ◽  
Spatial Databases ◽  
Robot Navigation ◽  
Spatial Relations ◽  
Spatial Representations ◽  
Language Understanding ◽  
Different Types ◽  
Qualitative Spatial Representation ◽  
Autonomous Robot Navigation ◽  
Future Work

AbstractRepresentation and reasoning with qualitative spatial relations is an important problem in artificial intelligence and has wide applications in the fields of geographic information system, computer vision, autonomous robot navigation, natural language understanding, spatial databases and so on. The reasons for this interest in using qualitative spatial relations include cognitive comprehensibility, efficiency and computational facility. This paper summarizes progress in qualitative spatial representation by describing key calculi representing different types of spatial relationships. The paper concludes with a discussion of current research and glimpse of future work.

scholarly journals Spatial Humanities 3.0: Qualitative Spatial Representation and Semantic Triples as New Means of Exploration of Complex Indigenous Spatial Representations in Sixteenth Century Early Colonial Mexican Maps

2019 ◽  
Vol 13 (1-2) ◽  
pp. 53-68 ◽  
Author(s):  
Patricia Murrieta-Flores ◽  
Mariana Favila-Vázquez ◽  
Aban Flores-Morán
Keyword(s):  
Spatial Representation ◽  
Spatial Information ◽  
Theoretical Perspective ◽  
Early Colonial ◽  
Starting Point ◽  
Model Complex ◽  
Alternative Means ◽  
Spatial Exploration ◽  
Qualitative Spatial Representation

For some time now, the field of Spatial Humanities has acknowledged the need for a system capable of the spatial exploration of historical and archaeological phenomena beyond Geographic Information Systems (GIS). This idea comes from the need to analyse holistically spatial information, including that which is not geographic (i.e. vague, symbolic and imaginary space). In addition, this need becomes more apparent when dealing with traditions that do not conform to the Modern/European/Cartesian conception of space in which GIS is rooted. This article, explores the use of Qualitative Spatial Representation (QSR) and Semantic Triples as possible alternative means to model complex and diverse expressions of spatial information, including social and symbolic conceptions in 16th century Mexican maps. Using as case study the map from the region of Atengo-Misquiahuala (Hidalgo) which combine the Mesoamerican and European traditions, we explore how these approaches might open new venues of research, potentially shedding light to long discussed and problematic Mesoamerican spatial categories. Focusing on a contained and partial example, we examine from a theoretical perspective and as a starting point, the possible future implementation of these approaches for historical and archaeological research.

Sign in / Sign up

Export Citation Format

Share Document