Neural Learning Methods TRIZ Cases
- Virtual Agent for Listing Platform Query Response
A virtual agent analyzes user profiles to generate relevant responses, reducing computational resource consumption during complex query processing.
- Conditional Neural Network Layer Weights via Latent Parameters
A conditional neural network layer determines layer weights using continuous latent parameters mapped from decision parameters.
- Transfer Learning Model Generation via Weight Pruning
Reconstructing pre-trained models and removing weights based on activation degrees to generate task-specific transfer learning models.
- Dripout Mitigates Overfitting in Neural Network Training
Dripout applies influence-attenuating masks to neural network interior nodes during training.