Methods › General › Adaptive Activation Functions
Adaptive Activation Functions
The archive attaches this collection's text per method and the copies differ: 2 distinct texts across 5 of the 13 methods here. All are shown, most-carried first (a tie goes to the text carrying Papers with Code's collection boilerplate, then to the longer text); no vote is taken between them.
Text 1, carried by 4 of 13 methods:
Adaptive or trainable activation functions are the functions with trainable parameters that are able to adapt (change, optimize) their shape and amplitude to the target dataset.
Text 2, carried by 1 of 13 methods:
Fuzzy logic is a form of many-valued logic in which the truth value of variables may be any real number between 0 and 1. It is employed to handle the concept of partial truth, where the truth value may range between completely true and completely false. By contrast, in Boolean logic, the truth values of variables may only be the integer values 0 or 1.
Methods
All 13 methods in this collection, most-tagged first. Year is the archive's introduced_year; the archive stores 2000 when it has none, shown here as “–”. Papers counts distinct papers the archive tags with the method. Click a heading to sort.
| SPLASH Simple Piecewise Linear and Adaptive with Symmetric Hinges | – | 7 |
| modReLU | – | 6 |
| NFN Neo-fuzzy-neuron | – | 5 |
| Rational Activation function | – | 3 |
| AHAF Adaptive Hybrid Activation Function | – | 2 |
| DELU | – | 2 |
| PAU Padé Activation Units | – | 2 |
| PELU Parametric Exponential Linear Unit | – | 2 |
| AGLU Adaptive Generalised Linear Unit | – | 1 |
| CosLU Cosine Linear Unit | – | 1 |
| NormLinComb Normalized Linear Combination of Activations | – | 1 |
| PMish Parametric Mish | – | 1 |
| ReLUN Rectified Linear Unit N | – | 1 |