Welcome to my homepage

Hinton, 1986, Learning Distributed Representations of Concepts 소스

2016-10-10 23:39:01

프로그래밍

/**

name: Kyungmin Lee

- Weight update rules were improved(?)

- print out the best distributed representations

- minor bug fix: valid set index

*/

#include <stdio.h>

#include <math.h>

#include <stdlib.h> /* srand, rand */

#define N_PERSON 24 //size of one-hot encoding of persons

#define N_RELATION 12 //size of one-hot encoding of relations

#define N_DEP 6 //size of distributed encoding of persons

#define N_DER 6 //size of distributed encoding of relations

#define N_TRAIN 100 //size of training set

#define N_VALID 4 //size of validation set

#define N_CENTRAL 12 //size of central layer

#define MAX_STR 20

#define NEGLIGIBLE_ERROR 0 //stop condition

#define MAX_ITER 1500 //stop condition, accoring to Hinton's paper 573 for reducing error to be negligible or 1500 for learing distributed encodings,

//#define DEBUG

const char names[N_PERSON][MAX_STR]

= { "Christopher", "Penelope", "Andrew", "Christine", "Margaret", "Arthur",

"Colin", "Roberto", "Maria", "Gina", "Emilio", "Alfonso",

"Lucia", "Marco", "Sophia", "Victoria", "James", "Charlotte",

"Pierro", "Francesca", "Angela", "Tomaso", "Jennifer", "Charles" };

const char relation[N_RELATION][MAX_STR]

= { "son", "daughter", "nephew", "niece",

"father", "mother", "uncle", "aunt",

"brother", "sister", "husband", "wife" };

float LR = 0.3f; //Learning Rate

float ALPHA = 0.3f; //Exponential decay factor

#define SLOPE_SIGMOID 1//slope of sigmoid functions

//5 layers: all layers are sigmoid ?

float outputPersonVec[N_TRAIN][N_PERSON];

float outputPersonDE[N_TRAIN][N_DEP];

float centralLayer[N_TRAIN][N_CENTRAL];

float inputDE[N_TRAIN][N_DEP + N_DER]; //The second layer has two groups

float min_VE = 100.0f;

//Training Set

int trainSet[N_TRAIN][3]; //0: Person1, 1: Relation, 2: Person2

//Validation Set

int validSet[N_VALID][3]; //0: Person1, 1: Relation, 2: Person2

//5 weight matrix

float w_pv2de[N_PERSON][N_DEP]; //weights Person Vector -> Person Distributed Encoding(Encoding)

float min_w_pv2de[N_PERSON][N_DEP];

float w_rv2de[N_RELATION][N_DER]; //weights Relation Vector -> Relation Distributed Encoding(Encoding)

float min_w_rv2de[N_RELATION][N_DER];

float w_de2cl[N_DEP + N_DER][N_CENTRAL]; //weights Distributed Encoding -> Central layer

float w_cl2pd[N_CENTRAL][N_DEP]; //weights Central layer -> Pe\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0rson Distributed Encoding

float w_pd2pv[N_DEP][N_PERSON]; //weights Person Distributed Encoding -> Person Vector(Decoding)

int iteration = 0;

inline float sigmoid(float z) { return (1 / (1 + exp(SLOPE_SIGMOID * -z)));}

inline float derivatives_sigmoid(float y) {return SLOPE_SIGMOID * y * (1 - y); }

//inline void idxToVec(int idx, float* vec, int n) { for (int i = 0; i < n; i++) vec[i] = idx == i ? 1.0f : 0.0f;}

bool isCorrect;

void InitData() {

//w_pv2de

for (int j = 0; j < N_DEP; j++)

for (int i = 0; i < N_PERSON; i++)

w_pv2de[i][j] = (int)((((rand() % 600 + 1) / 1000.0f) - 0.3) * 10) / 10.0f;

//w_rv2de

for (int j = N_DEP; j < N_DEP + N_DER; j++)

for (int i = 0; i < N_RELATION; i++)

w_rv2de[i][j - N_DEP] = (int)((((rand() % 600 + 1) / 1000.0f) - 0.3) * 10) / 10.0f;

//w_de2cl

for (int k = 0; k < N_CENTRAL; k++)

for (int j = 0; j < N_DEP + N_DER; j++)

w_de2cl[j][k] = (int)((((rand() % 600 + 1) / 1000.0f) - 0.3) * 10) / 10.0f;

//w_cl2pd

for (int l = 0; l < N_DEP; l++)

for (int k = 0; k < N_CENTRAL; k++)

w_cl2pd[k][l] = (int)((((rand() % 600 + 1) / 1000.0f) - 0.3) * 10) / 10.0f;

//w_pd2pv

for (int m = 0; m < N_PERSON; m++)

for (int l = 0; l < N_DEP; l++)

w_pd2pv[l][m] = (int)((((rand() % 600 + 1) / 1000.0f) - 0.3) * 10) / 10.0f;

}

float maxAverage = 0;

float minAverage = 0;

float FeedForward(int p1, int r, int p2, int c) {

//Person Vector -> Person Distributed Encoding(Encoding)

for (int j = 0; j < N_DEP; j++)

inputDE[c][j] = sigmoid(w_pv2de[p1][j]);

//Relation Vector -> Relation Distributed Encoding(Encoding)

for (int j = 0; j < N_DER; j++)

inputDE[c][j + N_DEP] = sigmoid(w_rv2de[r][j]);

//Distributed Encoding -> Central layer

for (int k = 0; k < N_CENTRAL; k++) {

float z_centralLayer = 0;

for (int j = 0; j < N_DEP + N_DER; j++) z_centralLayer += w_de2cl[j][k] * inputDE[c][j];

centralLayer[c][k\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0] = sigmoid(z_centralLayer);

}

//Central layer -> Person Distributed Encoding

for (int l = 0; l < N_DEP; l++) {

float z_outputPersonDE = 0;

for (int k = 0; k < N_CENTRAL; k++) z_outputPersonDE += w_cl2pd[k][l] * centralLayer[c][k];

outputPersonDE[c][l] = sigmoid(z_outputPersonDE);

}

//Person Distributed Encoding -> Person Vector(Decoding)

float error = 0;

int isBiggerThanHalf = 0;

float max = 0;

float min = 100000000000;

int maxIdx = -1;

for (int m = 0; m < N_PERSON; m++) {

float z_outputPersonVec = 0;

for (int l = 0; l < N_DEP; l++) z_outputPersonVec += w_pd2pv[l][m] * outputPersonDE[c][l];

outputPersonVec[c][m] = sigmoid(z_outputPersonVec);

//Error: { sum ( y - d )^2 } / 2

{

if (outputPersonVec[c][m] > max)

max = outputPersonVec[c][m]; maxIdx = m;

if (outputPersonVec[c][m] < min) min = outputPersonVec[c][m];

if (p2 == m) error += pow(outputPersonVec[c][m] - 1, 2);

else error += pow(outputPersonVec[c][m] - 0, 2);

if (p2 == m && outputPersonVec[c][m] >= 0.5) isBiggerThanHalf = 1;

else if (p2 == m && outputPersonVec[c][m] < 0.5) isBiggerThanHalf--;

else if (p2 != m && outputPersonVec[c][m] >= 0.5) isBiggerThanHalf--;

}

maxAverage += max;

minAverage += min;

isCorrect = isBiggerThanHalf == 1;

return error / 2.0f;

}

float p_d_w_pv2de[N_PERSON][N_DEP]; //delta of weights Person Vector -> Person Distributed Encoding(Encoding)

float p_d_w_rv2de[N_RELATION][N_DER]; //delta of weights Relation Vector -> Relation Distributed Encoding(Encoding)

float p_d_w_de2cl[N_DEP + N_DER][N_CENTRAL]; //delta of weights Distributed Encoding -> Central layer

float p_d_w_cl2pd[N_CENTRAL][N_DEP]; //delta of weights Central layer -> Person Distributed Encoding

float p_d_w_pd2pv[N_DEP][N_PERSON]; &nbs\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0p; //delta of weights Person Distributed Encoding -> Person Vector(Decoding)

void BackPropagation(bool isFirst) {

float d_w_pv2de[N_PERSON][N_DEP]; //delta of weights Person Vector -> Person Distributed Encoding(Encoding)

float d_w_rv2de[N_RELATION][N_DER]; //delta of weights Relation Vector -> Relation Distributed Encoding(Encoding)

float d_w_de2cl[N_DEP + N_DER][N_CENTRAL]; //delta of weights Distributed Encoding -> Central layer

float d_w_cl2pd[N_CENTRAL][N_DEP]; //delta of weights Central layer -> Person Distributed Encoding

float d_w_pd2pv[N_DEP][N_PERSON]; //delta of weights Person Distributed Encoding -> Person Vector(Decoding)

//Initialize deltas of weights

//w_pv2de

for (int j = 0; j < N_DEP; j++)

for (int i = 0; i < N_PERSON; i++) {

d_w_pv2de[i][j] = 0;

if (isFirst) p_d_w_pv2de[i][j] = 0;

}

//w_rv2de

for (int j = 0; j < N_DER; j++)

for (int i = 0; i < N_RELATION; i++) {

d_w_rv2de[i][j] = 0;

if(isFirst) p_d_w_rv2de[i][j] = 0;

}

//w_de2cl

for (int k = 0; k < N_CENTRAL; k++)

for (int j = 0; j < N_DEP + N_DER; j++) {

d_w_de2cl[j][k] = 0;

if (isFirst) p_d_w_de2cl[j][k] = 0;

}

//w_cl2pd

for (int l = 0; l < N_DEP; l++)

for (int k = 0; k < N_CENTRAL; k++) {

d_w_cl2pd[k][l] = 0;

if (isFirst) p_d_w_cl2pd[k][l] = 0;

}

//w_pd2pv

for (int m = 0; m < N_PERSON; m++)

for (int l = 0; l < N_DEP; l++) {

d_w_pd2pv[\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0l][m] = 0;

if (isFirst) p_d_w_pd2pv[l][m] = 0;

}

for (int c = 0; c < N_TRAIN; c++) {

float pEBypZm[N_PERSON];

float pEBypZl[N_DEP];

float pEBypZk[N_CENTRAL];

float pEBypZj[12];

//Output => P2 Destributed Encoding: d_w_pd2pv

for (int l = 0; l < N_DEP; l++) { //l

for (int m = 0; m < N_PERSON; m++) { //m

//ym -dm

float pEBypYm = trainSet[c][2] == m ? outputPersonVec[c][m] - 1 : outputPersonVec[c][m];

pEBypZm[m] = pEBypYm * derivatives_sigmoid(outputPersonVec[c][m]);

d_w_pd2pv[l][m] += pEBypZm[m] * outputPersonDE[c][l]; //accumulate

}

//P2 Distributed Encoding => Central: d_w_cl2pd

for (int l = 0; l < N_DEP; l++) {

pEBypZl[l] = 0;

for (int m = 0; m < N_PERSON; m++)

pEBypZl[l] += pEBypZm[m] * w_pd2pv[l][m];

pEBypZl[l] = pEBypZl[l] * derivatives_sigmoid(outputPersonDE[c][l]);

}

for (int k = 0; k < N_CENTRAL; k++)

for (int l = 0; l < N_DEP; l++)

d_w_cl2pd[k][l] += pEBypZl[l] * centralLayer[c][k];

//Central => Distributed Encoding: d_w_de2cl

for (int k = 0; k < N_CENTRAL; k++) {

pEBypZk[k] = 0;

for (int l = 0; l < N_DEP; l++)

pEBypZk[k] += pEBypZl[l] * w_cl2pd[k][l];

pEBypZk[k] *= derivatives_sigmoid(centralLayer[c][k]);

}

for (int j = 0; j < N_DEP + N_DER; j++)

for (int k = 0; k < N_CENTRAL; k++)

d_w_de2cl[j][k] += pEBypZk[k] * inputDE[c][j];

//Distributed Encoding => Relation Local: d_w_rv2de

for (int j = N_DEP; j < N_DEP + N_DER; j++) {

&nb\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0sp; pEBypZj[j] = 0;

for (int k = 0; k < N_CENTRAL; k++)

pEBypZj[j] += pEBypZk[k] * w_de2cl[j][k];

pEBypZj[j] *= derivatives_sigmoid(inputDE[c][j]);

}

for (int i = 0; i < N_RELATION; i++)

for (int j = N_DEP; j < N_DEP + N_DER; j++)

if (i == trainSet[c][1])

d_w_rv2de[i][j - N_DEP] += pEBypZj[j];// *inputRelationVec[c][i];

//Distributed Encoding => Person Local: d_w_pv2de

for (int j = 0; j < N_DEP; j++) {

pEBypZj[j] = 0;

for (int k = 0; k < N_CENTRAL; k++)

pEBypZj[j] += pEBypZk[k] * w_de2cl[j][k];

pEBypZj[j] *= derivatives_sigmoid(inputDE[c][j]);

}

for (int i = 0; i < N_PERSON; i++)

for (int j = 0; j < N_DEP; j++)

if (i == trainSet[c][0])

d_w_pv2de[i][j] += pEBypZj[j];// *inputPersonVec[c][i];

}//training iteration finished

//Update weights

//w_pv2de

for (int j = 0; j < N_DEP; j++)

for (int i = 0; i < N_PERSON; i++) {

w_pv2de[i][j] += -LR * d_w_pv2de[i][j] +ALPHA * p_d_w_pv2de[i][j];

p_d_w_pv2de[i][j] = -LR * d_w_pv2de[i][j] +ALPHA * p_d_w_pv2de[i][j];

}

//w_rv2de

for (int j = 0; j < N_DER; j++)

for (int i = 0; i < N_RELATION; i++) {

w_rv2de[i][j] += -LR * d_w_rv2de[i][j] +ALPHA * p_d_w_rv2de[i][j];

p_d_w_rv2de[i][j] = -LR * d_w_rv2de[i][j] +ALPHA * p_d_w_rv2de[i][j];

}

//w_\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0de2cl

for (int k = 0; k < N_CENTRAL; k++)

for (int j = 0; j < N_DEP + N_DER; j++) {

w_de2cl[j][k] += -LR * d_w_de2cl[j][k] +ALPHA * p_d_w_de2cl[j][k];

p_d_w_de2cl[j][k] = -LR * d_w_de2cl[j][k] +ALPHA * p_d_w_de2cl[j][k];

}

//w_cl2pd

for (int l = 0; l < N_DEP; l++)

for (int k = 0; k < N_CENTRAL; k++) {

w_cl2pd[k][l] += -LR * d_w_cl2pd[k][l] +ALPHA * p_d_w_cl2pd[k][l];

p_d_w_cl2pd[k][l] = -LR * d_w_cl2pd[k][l] +ALPHA * p_d_w_cl2pd[k][l];

}

//w_pd2pv

for (int m = 0; m < N_PERSON; m++)

for (int l = 0; l < N_DEP; l++) {

w_pd2pv[l][m] += -LR * d_w_pd2pv[l][m] +ALPHA * p_d_w_pd2pv[l][m];

p_d_w_pd2pv[l][m] = -LR * d_w_pd2pv[l][m] +ALPHA * p_d_w_pd2pv[l][m];

}

int maxCorrectN = 0;

int maxIteration = 0;

int main() {

//Local variables

float E;

float prevError = -1;

//Load Train, Valid Set

FILE *p = fopen("Hinton_FamilyTree.txt", "r");

if (p == NULL) {

printf("!!ERROR: No input file.");

return 1;

}

char temp[3];

for (int i = 0; i < 104; i++) {

if (i < 100) {

fscanf(p, "%s", temp); trainSet[i][0] = atoi(temp);

fscanf(p, "%s", temp); trainSet[i][1] = atoi(temp);

fscanf(p, "%s", temp); trainSet[i][2] = atoi(temp);

}else {

fscanf(p, "%s", temp); validSet[i-100][0] = atoi(temp);

fscanf(p, "%s", temp); validSet[i-100][1] = atoi(temp);

fscanf(p, "%s", temp); validSet[i-100][2] = atoi(temp);

}

fclose(p);

//Initialize weights

InitData();

//Training

float prevE = 0;

while(true) {

E = 0.0f;

int correctN = 0;

&n\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0bsp; maxAverage = 0;

minAverage = 0;

for (int c = 0; c < N_TRAIN;c++){

E += FeedForward(trainSet[c][0], trainSet[c][1], trainSet[c][2], c);

if (isCorrect) correctN++;

}

//printf("%.5f should be bigger than 0.5

", maxAverage / N_TRAIN);

//printf("%.5f should be smaller than 0.5

", minAverage / N_TRAIN);

int valCorrectN = 0;

float VE = 0.0f;

for (int c = 0; c < N_VALID; c++) {

VE += FeedForward(validSet[c][0], validSet[c][1], validSet[c][2], c);

if (isCorrect) {

valCorrectN++;

}

if(valCorrectN > maxCorrectN){

maxIteration = iteration;

maxCorrectN = valCorrectN;

}

printf("[%d] Error : %.5f, Learning Rate : %.5f, Decay Factor : %.5f, Train Correct %d out of %d, Valid Correct %d out of %d, Valid E %.5f

", iteration, E, LR, ALPHA, correctN, N_TRAIN, valCorrectN, N_VALID, VE);

if (iteration >= MAX_ITER) break;

BackPropagation(!iteration); // iteration

iteration++;

if (VE < 1.5) {

LR = 0.25;

ALPHA = 0.35;

}

prevE = VE;

if (VE < min_VE) {

min_VE = VE;

for (int i = 0; i < N_PERSON;i++)

for (int j = 0; j < N_DEP; j++) min_w_pv2de[i][j] = w_pv2de[i][j];

for (int i = 0; i < N_RELATION; i++)

for (int j = 0; j < N_\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0DER; j++) min_w_rv2de[i][j] = w_rv2de[i][j];

}

printf("Max correctN is %d out of %d, error %.5f at %d

", maxCorrectN, N_VALID, min_VE, maxIteration);

//Print out the elements of second layer to reproduce results of the Hinton's paper

//People

printf("Weights from the 24 input units for people at %d

", maxIteration);

printf("%s %s %s %s %s %s %s %s %s %s %s %s\t%s %s %s %s %s %s %s %s %s %s %s %s

",

names[0], names[2], names[5], names[16], names[23], names[6], names[1], names[3], names[15], names[22], names[4], names[17],

names[0], names[2], names[5], names[16], names[23], names[6], names[1], names[3], names[15], names[22], names[4], names[17]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_pv2de[0][3], min_w_pv2de[2][3], min_w_pv2de[5][3], min_w_pv2de[16][3], min_w_pv2de[23][3], min_w_pv2de[6][3], min_w_pv2de[1][3], min_w_pv2de[3][3], min_w_pv2de[15][3], min_w_pv2de[22][3], min_w_pv2de[4][3], min_w_pv2de[17][3],

min_w_pv2de[0][0], min_w_pv2de[2][0], min_w_pv2de[5][0], min_w_pv2de[16][0], min_w_pv2de[23][0], min_w_pv2de[6][0], min_w_pv2de[1][0], min_w_pv2de[3][0], min_w_pv2de[15][0], min_w_pv2de[22][0], min_w_pv2de[4][0], min_w_pv2de[17][0]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_pv2de[7][3], min_w_pv2de[18][3], min_w_pv2de[10][3], min_w_pv2de[13][3], min_w_pv2de[21][3], min_w_pv2de[11][3], min_w_pv2de[8][3], min_w_pv2de[19][3], min_w_pv2de[12][3], min_w_pv2de[20][3], min_w_pv2de[9][3], min_w_pv2de[14][3],

min_w_pv2de[7][0], min_w_pv2de[18][0], min_w_pv2de[10][0], min_w_pv2de[13][0], min_w_pv2de[21][0], min_w_pv2de[11][0], min_w_pv2de[8][0], min_w_pv2de[19][0], min_w_pv2de[12][0], min_w_pv2de[20][0], min_w_pv2de[9][0], min_w_pv2de[14][0]);

printf("

");

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_pv2de[0][4], min_w_pv2de[2][4], min_w_pv2de[5][4], min_w_pv2de[16][4], min_w_pv2de[23][4], min_w_pv2de[6][4], min_w_pv2de[1][4], min_w_pv2de[3][4], min_w_pv2de[15][4], min_w_pv2de[22][4], min_w_pv2de[4][4], min_w_pv2de[17][4],

min_w_pv2de[0][1], min_w_pv2de[2][1], min_w_pv2de[5][1], min_w_pv2de[16][1], min_w_pv2de[23][1], min_w_pv2de[6][1], min_w_pv2de[1][1], min_w_pv2de[3][1], min_w_pv2de[15][1], min_w_pv2de[22][1], min_w_pv2de[4][1], min_w_pv2de[17][1]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_pv2de[7][4], min_w_pv2de[18][4], min_w_pv2de[10][3], min_w_pv2de[13][4], min_w_pv2de[21][4], min_w_pv2de[11][4], min_w_pv2de[8][4], min_w_pv2de[19][4], min_w_pv2de[12][4], min_w_pv2de[20][4], min_w_pv2de[9][4], min_w_pv2de[14][4],

min_w_pv2de[7][1], min_w_pv2de[18][1], min_w_pv2de[10][3], min_w_pv2de[13][1], min_w_pv2de[21][1], min_w_pv2de[11][1], min_w_pv2de[8][1], min_w_pv2de[19][1], min_w_pv2de[12][1], min_w_pv2de[20][1], min_w_pv2de[9][1], min_w_pv2de[14][1]);

printf("

");

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

", min_w_pv2de[0][5], min_w_pv2de[2][5],\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0 min_w_pv2de[5][5], min_w_pv2de[16][5], min_w_pv2de[23][5], min_w_pv2de[6][5], min_w_pv2de[1][5], min_w_pv2de[3][5], min_w_pv2de[15][5], min_w_pv2de[22][5], min_w_pv2de[4][5], min_w_pv2de[17][5],

min_w_pv2de[0][2], min_w_pv2de[2][2], min_w_pv2de[5][2], min_w_pv2de[16][2], min_w_pv2de[23][2], min_w_pv2de[6][2], min_w_pv2de[1][2], min_w_pv2de[3][2], min_w_pv2de[15][2], min_w_pv2de[22][2], min_w_pv2de[4][2], min_w_pv2de[17][2]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_pv2de[7][5], min_w_pv2de[18][5], min_w_pv2de[10][5], min_w_pv2de[13][5], min_w_pv2de[21][5], min_w_pv2de[11][5], min_w_pv2de[8][5], min_w_pv2de[19][5], min_w_pv2de[12][5], min_w_pv2de[20][5], min_w_pv2de[9][5], min_w_pv2de[14][5],

min_w_pv2de[7][2], min_w_pv2de[18][2], min_w_pv2de[10][2], min_w_pv2de[13][2], min_w_pv2de[21][2], min_w_pv2de[11][2], min_w_pv2de[8][2], min_w_pv2de[19][2], min_w_pv2de[12][2], min_w_pv2de[20][2], min_w_pv2de[9][2], min_w_pv2de[14][2]);

printf("%s %s %s %s %s %s %s %s %s %s %s %s\t%s %s %s %s %s %s %s %s %s %s %s %s

",

names[7], names[18], names[10], names[13], names[21], names[11], names[8], names[19], names[12], names[20], names[9], names[14],

names[7], names[18], names[10], names[13], names[21], names[11], names[8], names[19], names[12], names[20], names[9], names[14]);

printf("

");

printf("Weights from the 12 input units for relations at %d

", maxIteration);

printf("

%s %s %s %s %s %s\t%s %s %s %s %s %s\t%s %s %s %s %s %s

",

relation[8], relation[11], relation[0], relation[1], relation[4], relation[5], relation[8], relation[11], relation[0], relation[1], relation[4], relation[5], relation[8], relation[11], relation[0], relation[1], relation[4], relation[5]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f

",

min_w_rv2de[8][4], min_w_rv2de[11][4], min_w_rv2de[0][4], min_w_rv2de[1][4], min_w_rv2de[4][4], min_w_rv2de[5][4],

min_w_rv2de[8][2], min_w_rv2de[11][2], min_w_rv2de[0][2], min_w_rv2de[1][2], min_w_rv2de[4][2], min_w_rv2de[5][2],

min_w_rv2de[8][0], min_w_rv2de[11][0], min_w_rv2de[0][0], min_w_rv2de[1][0], min_w_rv2de[4][0], min_w_rv2de[5][0]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_rv2de[10][4], min_w_rv2de[9][4], min_w_rv2de[2][4], min_w_rv2de[3][4], min_w_rv2de[6][4], min_w_rv2de[7][4],

min_w_rv2de[10][2], min_w_rv2de[9][2], min_w_rv2de[2][2], min_w_rv2de[3][2], mi\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0n_w_rv2de[6][2], min_w_rv2de[7][2],

min_w_rv2de[10][0], min_w_rv2de[9][0], min_w_rv2de[2][0], min_w_rv2de[3][0], min_w_rv2de[6][0], min_w_rv2de[7][0]);

printf("

");

printf("%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f

",

min_w_rv2de[8][5], min_w_rv2de[11][5], min_w_rv2de[0][5], min_w_rv2de[1][5], min_w_rv2de[4][5], min_w_rv2de[5][5],

min_w_rv2de[8][3], min_w_rv2de[11][3], min_w_rv2de[0][3], min_w_rv2de[1][3], min_w_rv2de[4][3], min_w_rv2de[5][3],

min_w_rv2de[8][1], min_w_rv2de[11][1], min_w_rv2de[0][1], min_w_rv2de[1][1], min_w_rv2de[4][1], min_w_rv2de[5][1]);

printf("%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f\t%.2f %.2f %.2f %.2f %.2f %.2f

"

, min_w_rv2de[10][5], min_w_rv2de[9][5], min_w_rv2de[2][5], min_w_rv2de[3][5], min_w_rv2de[6][5], min_w_rv2de[7][5],

min_w_rv2de[10][3], min_w_rv2de[9][3], min_w_rv2de[2][3], min_w_rv2de[3][3], min_w_rv2de[6][3], min_w_rv2de[7][3],

min_w_rv2de[10][1], min_w_rv2de[9][1], min_w_rv2de[2][1], min_w_rv2de[3][1], min_w_rv2de[6][1], min_w_rv2de[7][1]);

printf("%s %s %s %s %s %s\t%s %s %s %s %s %s\t%s %s %s %s %s %s

",

relation[10], relation[9], relation[2], relation[3], relation[6], relation[7], relation[10], relation[9], relation[2], relation[3], relation[6], relation[7], relation[10], relation[9], relation[2], relation[3], relation[6], relation[7]);

}

▼ more

Hinton, 1986, Learning Distributed Representations of Concepts 구현

2016-10-10 22:34:33

프로그래밍

The purpose of this paper(Hinton '86) is teaching NN to understand relationships between P1 and P2.

Family Tree

The family tree above can be expressed by a set of tripletts.

Person1 Relation Person2

Christopher wife Penelope

Arthur wife Margaret

James wife Victoria

Andrew wife Christine

Charles wife Jennifer

Roberto wife Maria

Pierro wife Francesca

The family tree contains 104 triplett.(FamilyTree.xlsx)

Neural Network Structure

Feed forward

There are not bias nodes.

$^{x_{j}}=\sum _i y_iw_{ji}$

Non-linear function(sigmoid function)

$y_j=\frac{1}{1+e^{-x_j}}$

The units are arrainged in layers with a layer of input uints at the bottom, any number of intermediate layers, and a layer of output uints at the top. There no feedback connections.

Back propagation

Squared residual errors, no bias node, Batch mode

acceleration medthod : delta W(t-1)

$\Delta w(t)=-\varepsilon \frac{\partial E}{\partial w(t)}+\alpha \Delta w(t-1)$

t is incremented by 1 for each sweep through the whole set of input-output cases, and alpha is an exponential decay factor between 0 and 1 that determines the relative contribution of the current gradient and ealier gradients on the weight change.

The results.

Weights from the 24 input units for people

Weights from the 12 input units for relations

My model got 2 out of 4 test cases,

wheree "correct" means that the output unit corresponding to the right answer had an activity level above 0.5, and all the other output units were below 0.5.

▼ more

수식은 잘 나오나요?

2016-10-10 19:40:29

홈페이지

내가 아는 BP

$\frac{\partial z }{\partial x_i}= \sum _j \frac{\partial z }{\partial y_j} \frac{\partial y_j }{\partial x_i}$

▼ more

망각력을 보여준 통계학시험

2016-10-10 18:50:03

일상

어마어마한 망각력과 헷갈림들;;

그래도 하루는 뚫어져라 보긴했는데..

아직 남은 중간고사 2 를 기대할 뿐이다.

그리고 연습문제는 이번처럼 반드시 모두 풀어봐야겠다. 홀수만이라도

안풀어봤으면 그나마 더 당황했을 것 같으니..

그리고 앞으로는 3 일은 공부를 해야겠다.

하루는 문제풀이 하기에도 부족하다는 것을 몸소 깨달은 시험이었다.

어쨌거나 시험을 핑계로 잘만큼 자고 나니

여러 의미로 개운하다.

할만큼 해보고 결과를 보고 그다음에 생각하자

어찌 되었던 쓸데없는 생각은 덜하게 되지 않았는가

▼ more

Lee, Kyungmin'sPersonal Blog

Lee, Kyungmin's
Personal Blog