Researchers have introduced PIE-APT, a novel framework for abductive planning over Temporal Dynamic Knowledge Graphs (TDKGs). This system operates within open-world environments and addresses challenges like decidability issues and the Ramification Problem inherent in existing action formalisms. PIE-APT utilizes a unified approach with two modules, PIE-Abducer and PIE-APT, which work natively with the SROIQ Description Logic and OWL to model state transitions and actions. The framework employs an incremental reasoner to maintain decidability and bypasses the Ramification Problem by representing actions in OWL. It also circumvents traditional Minimal Hitting Set enumeration for incomplete knowledge by synthesizing missing premises through direct refutation consequences, outperforming classical planners and a baseline in abductive enrichment. AI
IMPACT Introduces a new method for AI planning that could improve reasoning capabilities in complex, open-world environments.
RANK_REASON The cluster describes a new academic paper detailing a novel framework for AI planning. [lever_c_demoted from research: ic=1 ai=1.0]
- Amir Hossein Sharafi
- A* search algorithm
- Minimal Hitting Set
- PIE-Abducer
- PIE-APT
- SROIQ Description Logic
- Temporal Dynamic Knowledge Graphs
- Temporal Projection
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
