Research Article
Compressive Strength Evaluation of Fiber-Reinforced High-Strength Self-Compacting Concrete with Artificial Intelligence
Table 3
Performance of the ANN model with different learning algorithms.
| No. | Algorithm | Details | TRN | TES | VAL | EPC |
| 1 | Trainrp | Resilient backpropagation | 70.8 | 42.5 | 25.3 | 6 | 2 | Trainlm | Levenberg-Marquardt backpropagation | 8.29 | 31.6 | 21.2 | 6 | 3 | Traincgp | Conjugate gradient backpropagation with Polack–Ribiere updates | 131 | 43.9 | 105 | 6 | 4 | Traincgb | Conjugate gradient backpropagation with Power-Beale restarts | 71.9 | 38.2 | 76.9 | 6 | 5 | Trainbfg | BFGS quasi-Newton backpropagation | 78.4 | 36.0 | 28.9 | 7 | 6 | Trainoss | One-step secant backpropagation | 93.6 | 88.0 | 36.9 | 6 | 7 | Traincgf | Conjugate gradient backpropagation with Fletcher–Reeves updates | 69.9 | 56.5 | 38.8 | 7 | 8 | Traingda | Gradient descent with adaptive learning rate backpropagation | 192 | 128 | 113 | 20 |
|
|