Class | Description |
---|---|
BestPairwiseAlignmentKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = softmax(BK(x_1, y_1) \cdot BK(x_2, y_2) , BK(x_1, y_2) \cdot BK(x_2, y_1))\) |
KernelOnPairs |
It is a kernel operating on ExamplePairs applying a simpler kernel to the pair elements
|
PairwiseProductKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = BK(x_1, y_1) \cdot BK(x_2, y_2) + BK(x_1, y_2) \cdot BK(x_2, y_1)\) where BK is another kernel the kernel on pairs relies on. |
PairwiseSumKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = BK(x_1, y_1) + BK(x_2, y_2) + BK(x_1, y_2) + BK(x_2, y_1)\) where BK is another kernel the kernel on pairs relies on. |
PreferenceKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = BK(x_1, y_1) + BK(x_2, y_2) - BK(x_1, y_2) - BK(x_2, y_1)\) where BK is another kernel the preference kernel relies on. |
UncrossedPairwiseProductKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = BK(x_1, y_1) \cdot BK(x_2, y_2)\) where BK is another kernel the kernel on pairs relies on. |
UncrossedPairwiseSumKernel |
It is a kernel operating on ExamplePairs applying the following formula:
\(K( < x_1, x_2 >, < y_1,y_2 > ) = BK(x_1, y_1) + BK(x_2, y_2)\) where BK is another kernel the kernel on pairs relies on. |
Copyright © 2018 Semantic Analytics Group @ Uniroma2. All rights reserved.