Statements (60)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:neural_networks
|
| gptkbp:analyzes |
Decision boundary
|
| gptkbp:based_on |
Linear threshold unit
|
| gptkbp:can |
Linearly separable functions
|
| gptkbp:can_be_combined_with |
Other algorithms
|
| gptkbp:can_be_extended_by |
Multi-layer perceptron
|
| gptkbp:consists_of |
Weights
Activation function Input layer |
| gptkbp:developed_by |
gptkb:Frank_Rosenblatt
|
| gptkbp:has_limitations |
Overfitting
Cannot solve XOR problem |
| https://www.w3.org/2000/01/rdf-schema#label |
Perceptron
|
| gptkbp:improves |
Regularization techniques
|
| gptkbp:input_output |
Binary output
|
| gptkbp:inspired_by |
Biological neurons
|
| gptkbp:introduced_in |
gptkb:1958
|
| gptkbp:is_a |
Single-layer neural network
|
| gptkbp:is_described_as |
Learning theory
|
| gptkbp:is_evaluated_by |
Accuracy
Precision Recall F1 score Confusion matrix Test set |
| gptkbp:is_implemented_in |
gptkb:Tensor_Flow
gptkb:Py_Torch Various programming languages |
| gptkbp:is_influenced_by |
Hebbian learning
|
| gptkbp:is_part_of |
gptkb:Artificial_Intelligence
Predictive modeling Neural network architecture Supervised learning algorithms |
| gptkbp:is_related_to |
Cognitive science
Deep learning Machine learning Feature extraction Statistical learning theory |
| gptkbp:is_trained_in |
Gradient descent
|
| gptkbp:is_used_in |
gptkb:robotics
Data mining Natural language processing Speech recognition Anomaly detection Medical diagnosis Pattern recognition Game AI Recommendation systems Financial forecasting Image recognition Time series prediction |
| gptkbp:requires |
Training data
|
| gptkbp:training |
Supervised learning
Stochastic gradient descent Batch learning Backpropagation (in multi-layer perceptrons) |
| gptkbp:used_for |
Binary classification
|
| gptkbp:uses |
Activation threshold
|
| gptkbp:bfsParent |
gptkb:Ken_Mc_Culloch
|
| gptkbp:bfsLayer |
6
|