AIRIS (legacy duplicate)

Responsible: Berick Cook

GitHub / Demos:

• https://github.com/singnet/AIRIS-scripts/
• AIRIS Minecraft Agent
• AIRIS talk at AGI 24 in Seattle
• 2023 AIRIS demo(pre-SNET)
• "Technical Tuesday" livestream demo of the AIRIS Minecraft agent

Papers:

• Autonomous Intelligent Reinforcement Inferred Symbolism

Description:

AIRIS is a causal machine learning system designed to overcome the opacity and data-inefficiency of traditional deep reinforcement learning. Rather than ingesting massive datasets to approximate statistical correlations, AIRIS functions as a causal reasoner that actively constructs a deterministic model of its environment through direct interaction. This yields exceptional transparency, encoding decision logic in explicit, auditable rules rather than inscrutable neural weights.

The system has proven its capabilities in voxel-based environments like Minecraft, where it operates without pre-training. By observing the direct consequences of its actions (e.g., "walking into lava causes damage," "dirt blocks can be stacked"), AIRIS builds a dynamic knowledge base of causal rewrite rules. It uses these rules to run internal simulations and plan complex paths. Crucially, the system is self-correcting: when a prediction fails, AIRIS instantly isolates the error and updates its rule set, effectively applying the scientific method to autonomous navigation.

Within the context of Hyperon, AIRIS serves as a mechanism for causal learning, translating raw sensory data into the structured symbolic knowledge of the Atomspace. These grounded atoms become raw material for higher-level algorithms like PLN and MOSES.

Roadmap:

• Develop a "generalized" AIRIS that can accept any type of data from any domain
• Build the public API infrastructure for the generalized AIRIS
• Create demos of AIRIS operating in various domains