1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
|
#ifndef _CUBIASED_LINEARITY_H_
#define _CUBIASED_LINEARITY_H_
#include "cuComponent.h"
#include "cumatrix.h"
#include "Matrix.h"
#include "Vector.h"
namespace TNet {
/**
* \brief CuBiasedLinearity summation function
*
* \ingroup CuNNUpdatable
* Implements forward pass: \f[ Y_j=\Sigma_{i=0}^{i=N-1}w_ij X_i +{\beta}_j \f]
* Error propagation: \f[ E_i = \Sigma_{i=0}^{i=N-1} w_ij e_j \f]
*
* Weight adjustion: \f[ W_{ij} = (1-D)(w_{ij} - \alpha(1-\mu)x_i e_j - \mu \Delta) \f]
* and fot bias: \f[ {\Beta}_i = {\beta}_i - \alpha(1-\mu)e_i - \mu \Delta \f]
* where
* - D for weight decay => penalizing large weight
* - \f$ \alpha \f$ for learning rate
* - \f$ \mu \f$ for momentum => avoiding oscillation
*/
class CuBiasedLinearity : public CuUpdatableComponent
{
public:
CuBiasedLinearity(size_t nInputs, size_t nOutputs, CuComponent *pPred);
~CuBiasedLinearity();
ComponentType GetType() const;
const char* GetName() const;
void PropagateFnc(const CuMatrix<BaseFloat>& X, CuMatrix<BaseFloat>& Y);
void BackpropagateFnc(const CuMatrix<BaseFloat>& X, CuMatrix<BaseFloat>& Y);
void Update();
void ReadFromStream(std::istream& rIn);
void WriteToStream(std::ostream& rOut);
protected:
CuMatrix<BaseFloat> mLinearity; ///< Matrix with neuron weights
CuVector<BaseFloat> mBias; ///< Vector with biases
CuMatrix<BaseFloat> mLinearityCorrection; ///< Matrix for linearity updates
CuVector<BaseFloat> mBiasCorrection; ///< Vector for bias updates
};
////////////////////////////////////////////////////////////////////////////
// INLINE FUNCTIONS
// CuBiasedLinearity::
inline
CuBiasedLinearity::
CuBiasedLinearity(size_t nInputs, size_t nOutputs, CuComponent *pPred)
: CuUpdatableComponent(nInputs, nOutputs, pPred),
mLinearity(nInputs,nOutputs), mBias(nOutputs),
mLinearityCorrection(nInputs,nOutputs), mBiasCorrection(nOutputs)
{
mLinearityCorrection.SetConst(0.0);
mBiasCorrection.SetConst(0.0);
}
inline
CuBiasedLinearity::
~CuBiasedLinearity()
{ }
inline CuComponent::ComponentType
CuBiasedLinearity::
GetType() const
{
return CuComponent::BIASED_LINEARITY;
}
inline const char*
CuBiasedLinearity::
GetName() const
{
return "<biasedlinearity>";
}
} //namespace
#endif
|
