Artificial Life TRIZ Cases
- Game AI Policy Training Using Virtual Physiological Feedback
Virtual physiological signals guide reinforcement learning so game agents act more like human players with lower computational burden.
- AI Presenter Switching for Broadcast Hindrance in Web Conferences
Monitored noise, bandwidth, and computing limits trigger AI switching of screen-sharing control to keep web conference presentations clear.
- Neural Network Training Using a Parallel Teacher Model
A second neural network with a different architecture trains the operational model locally, cutting test-cycle time and data transfer in safety-critical systems.
- Virtual Agent Training Using Ranked Optimal Utterances
Conversation logs are mined and scored by syntactic and semantic similarity to automate virtual agent training phrase selection.
- AI Agent Orchestration for Context-Aware Enterprise Intervention
A control agent detects enterprise trigger events, deploys the right AI agents, and applies governance policies for accurate, efficient intervention.
- AI Agent Evaluation Scoring for Trustworthy Enterprise Deployment
Weighted scoring across behavior and trustworthiness helps enterprises evaluate AI agents in real conditions and trigger deployment intervention.
- RF Analog Circuit Parameter Generation Using GNN and Reinforcement Learning
A GNN-guided reinforcement learning agent automates RF analog circuit parameter tuning to meet target specs with less manual effort.
- Digital Twin Transfer Learning Selection for Low-Data Training
A transfer learning repository reuses prior digital twin training results to cut data transfer and training cost while preserving model quality.
- Fuzzy-Inference Task Allocation for Multi-Agent Sub-Task Conflicts
Fuzzy inference and dual-time-scale learning improve multi-agent sub-task allocation accuracy and efficiency in complex cooperative tasks.
- Wafer Slot Allocation Using Reinforcement Learning in Batch Tools
Reinforcement learning assigns wafer-specific batch slots from characteristic and history data to cut defects without slowing process timing.
- Virtual Avatar Interaction for Intelligent Digital Twin Communication
An AI avatar interprets digital twin state data and triggers context-aware user interaction, reducing expertise barriers in metaverse scenarios.
- Dynamic AI Chatbot Persona for Real-Time Response Customization
Real-time persona tuning lets an AI chatbot adjust tone, formality, and response speed to match user preferences and improve interaction quality.