AI & softwareResearch findings
Development of FNO-Based Data-Efficient Plasma Field Prediction Model
Original published 2026.11.11 13:00 KSTAdded to Blackdeck 2026.11.16 03:27 KST
What has changed?
FNO achieves the lowest or joint-lowest training data requirement in 1D and 2D CCP simulations
Key points
- CCP simulations are costly across operating conditions
- FNO requires the fewest reference simulations to achieve 1% error
- FNO demonstrates superior data efficiency in both 1D and 2D benchmarks
Connect it to your work
Reaches 1% error in 1D with 50 simulations; 2% error in 2D with 200 simulations
What remains unproven
The abstract reports simulation results; real-world performance is not established by this input.
Originals used for this note
Links open the original publisher in a new tab.