MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611116380 A) filed by Chandigarh University on September 29, 2026, for Causal Inconsistency Resolution Engine For Learning Systems Operating Under Partial And Evolving Truth Conditions.
Inventor includes Siddharth Arora.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention discloses a causal inconsistency resolution engine for learning systems operating under partial, evolving, or fragmented truth conditions. The engine 5 provides a dedicated software-based reasoning layer configured to continuously monitor, detect, and resolve contradictions among active causal relationships generated by one or more learning or reasoning models. The engine evaluates causal validity at runtime based on contextual relevance, temporal consistency, and confidence decay metrics. Upon detection of conflicting causal chains, the engine dynamically reweights, 10 compartmentalizes, or temporarily suspends the conflicting causal chains to prevent contradictory reasoning from influencing downstream inference, decision-making, or actuation processes. The engine further supports concurrent maintenance of multiple partially valid causal representations and enables stable operation without requiring modification of underlying learning architectures. The engine is model-agnostic and is 15 configured for embedding within software systems, autonomous machines, or embedded devices, thereby improving reliability, coherence, and operational stability under uncertainty and continuously changing conditions.
Disclaimer: Curated by HT Syndication.
