geo-infer-cog
Cognitive modeling for geospatial agents including attention, memory, and trust. Use when implementing spatial attention mechanisms, working memory for geographic contexts, or trust dynamics in multi-agent geospatial systems.
Resources
8Install
npx skillscat add activeinferenceinstitute/geo-infer/geo-infer-cog Install via the SkillsCat registry.
GEO-INFER-COG provides cognitive modeling capabilities for geospatial agents, including spatial attention mechanisms, working memory for geographic contexts, and trust dynamics in multi-agent systems. It addresses the need to model agent cognition when reasoning about spatial environments, such as selectively focusing on geographic regions or managing collaborative trust. Developers should use it when building geospatial agents that require cognitive architectures, particularly alongside GEO-INFER-ACT for active inference.
GEO-INFER-COG
Instructions
Core Capabilities
- Spatial attention: Selective attention to geographic regions
- Working memory: Context-dependent spatial memory buffers
- Trust dynamics: Agent trust modeling in collaborative environments
- Cognitive load: Resource allocation for spatial reasoning
- Mental models: Internal representations of spatial environments
Key Imports
from geo_infer_cog.core.attention import SpatialAttention
from geo_infer_cog.core.memory import WorkingMemory
from geo_infer_cog.core.trust import TrustModelExamples
from geo_infer_cog.core.attention import SpatialAttention
attention = SpatialAttention(resolution=7)
salience_map = attention.compute(observations, priors)
focus_region = attention.select_focus(salience_map)Guidelines
- Currently Alpha status — attention and memory models in development
Integrations
- Integrates with ACT for Active Inference-based cognitive architectures
- Test:
uv run python -m pytest GEO-INFER-COG/tests/ -v
Categories
Install
npx skillscat add activeinferenceinstitute/geo-infer/geo-infer-cog Install via the SkillsCat registry.