単純な強化学習実装シナリオ−学習オートマチックに基づくリンク予測モデル

学習を強化する.

    （  ：Reinforcement learning，  RL）           ，           ，           。                 ，                     ，          ，               。         ，             ，     、   、   、   、    、       、    、         。               ，       “      ”（approximate dynamic programming，ADP）。                ，                    ，           。         ，                        。

        ，               （MDP），                        。                  ，       MDP   ，                MDP。

                    ，            /   ，              。             ，     （      ）   （    ）      。      “  -  ”   ，           MDP      。

学習オートマトン

                       ，                    。               ，                        。

         ，                。          π，             。                。

リンク予測

        (Link Prediction)                                                。             （exist yet unknown links）            （future links）   。                           。

学習オートマチックに基づくリンク予測モデルの実現
import numpy as npimport timefrom random import choice import pandas as pdimport os
共通隣接指標を計算する方法を定義する
define some functions to calculate some baseline index
'''def Cn(MatrixAdjacency):

Matrix_similarity = np.dot(MatrixAdjacency,MatrixAdjacency)
return Matrix_similarity

'''
Jaccard類似性指標の計算
def Jaccavrd(MatrixAdjacency_Train):

Matrix_similarity = np.dot(MatrixAdjacency_Train,MatrixAdjacency_Train)

deg_row = sum(MatrixAdjacency_Train)
deg_row.shape = (deg_row.shape[0],1)
deg_row_T = deg_row.T
tempdeg = deg_row + deg_row_T
temp = tempdeg - Matrix_similarity

Matrix_similarity = Matrix_similarity / temp
return Matrix_similarity

Salton指標の計算方法の定義
def Salton_Cal(MatrixAdjacency_Train):

similarity = np.dot(MatrixAdjacency_Train,MatrixAdjacency_Train)

deg_row = sum(MatrixAdjacency_Train)
deg_row.shape = (deg_row.shape[0],1)
deg_row_T = deg_row.T
tempdeg = np.dot(deg_row,deg_row_T)
temp = np.sqrt(tempdeg)

np.seterr(divide='ignore', invalid='ignore')
Matrix_similarity = np.nan_to_num(similarity / temp)

print np.isnan(Matrix_similarity)
Matrix_similarity = np.nan_to_num(Matrix_similarity)
print np.isnan(Matrix_similarity)

return Matrix_similarity

Katz 1指標の計算方法の定義
def Katz_Cal(MatrixAdjacency):

#α  
Parameter = 0.01
Matrix_EYE = np.eye(MatrixAdjacency.shape[0])
Temp = Matrix_EYE - MatrixAdjacency * Parameter

Matrix_similarity = np.linalg.inv(Temp)

Matrix_similarity = Matrix_similarity - Matrix_EYE
return Matrix_similarity

ローカルパスLP類似性指標を計算する方法を定義する
'''def LP_Cal(MatrixAdjacency):

Matrix_similarity = np.dot(MatrixAdjacency,MatrixAdjacency)

Parameter = 0.05
Matrix_LP = np.dot(np.dot(MatrixAdjacency,MatrixAdjacency),MatrixAdjacency) * Parameter

Matrix_similarity = np.dot(Matrix_similarity,Matrix_LP)
return Matrix_similarity

'''
リソース割当類似性指標の計算
def RA(MatrixAdjacency_Train):

RA_Train = sum(MatrixAdjacency_Train)
RA_Train.shape = (RA_Train.shape[0],1)
MatrixAdjacency_Train_Log = MatrixAdjacency_Train / RA_Train
MatrixAdjacency_Train_Log = np.nan_to_num(MatrixAdjacency_Train_Log)

Matrix_similarity = np.dot(MatrixAdjacency_Train,MatrixAdjacency_Train_Log)
return Matrix_similarity

ランダム環境1:アクティブなノードペア
def RandomEnviromentForActive(MatrixAdjacency,i,j):

Index = np.random.randint(1, 5)
print(Index)
global IndexName
if Index == 1:
    IndexName = '      ：Jaccard Index'
    print(IndexName)
    similarity_matrix = Jaccavrd(MatrixAdjacency)
    similarity = similarity_matrix[i,j]
elif Index == 2:
    IndexName = '      ：Salton Index'
    print(IndexName)
    similarity_matrix = Salton_Cal(MatrixAdjacency)
    similarity = similarity_matrix[i,j]
elif Index == 3:
    IndexName = '      ：Katz Index'
    print(IndexName)
    similarity_matrix = Katz_Cal(MatrixAdjacency)
    similarity = similarity_matrix[i,j]
else index == 4:
    IndexName = '      ：RA Index'
    print(IndexName)
    similarity_matrix = RA(MatrixAdjacency)
    similarity = similarity_matrix[i,j]
return similarity 

ランダム環境2:主に非アクティブなノードペア
def RandomEnviromentForNonActive():

Action = np.random.randint(1, 4)
if Action == 1:
    ActionName = 'ID3'
    similarity_matrix = ID3_Cal(MatrixAdjacency)
    #similarity = similarity_matrix[i,j]
elif Action == 2:
    ActionName = 'CART'
    similarity_matrix = Cart_Cal(MatrixAdjacency)
    #similarity = similarity_matrix[i,j]
elif Action == 3:
    ActionName = 'C4.5'
    similarity_matrix = C4_Cal(MatrixAdjacency)
    #similarity = similarity_matrix[i,j]
return similarity

学習オートマトンの構築(To Construct the agent)
def ContructionAgent(filepath,n1,n2):

f = open(filepath)
lines = f.readlines()
A = np.zeros((50, 50), dtype=float)
A_row = 0
for line in lines:
    list = line.strip('
').split(' ')
    A[A_row:] = list[0:50]
    A_row += 1

#    p1 p2
a = 0.05
b = 0.01
p1 =0.5
p2 =0.5
Action = 1
#        1      ‘Yes’, 2    ‘No’
for i in range(1):

    #         global Action
    #      （the threashhold_value of similarity）
    if (p1 >= p2):
        Action = 1
    else:
        Action = 2
    print('      ：' + str(Action))
    threshhold_value = 0.3
    similarity = RandomEnviromentForActive(A, n1, n2)
    # p1    1'Yes'      ，p2    2'No'      
    #          ‘Yes’，          
    if (similarity > threshhold_value) and (Action == 1):
        p1 = p1 + a * (1 - p1)
        p2 = 1-p1
       # p2 = (1 - a) * p2
    #          'No',          
    elif (similarity < threshhold_value) and (Action == 2):
        p2 = (1-a)*p2
        p1 = 1-p2
       # p1 = (1 - a) * p1
    #          ‘Yes’，         
    elif (similarity < threshhold_value) and (Action == 1):
        p2 = 1-b*p2
        p1 = 1-p2
        #p2 = 1 - b * p2

    #          ‘No’，         
    elif (similarity > threshhold_value) and (Action == 2):
        p1 = b + (1 - b) * (1 - p1)
        p2 = 1-p1
       # p1 = 1 - b * p1

if (p1 >= p2):
    print('          :Yes')
else:
    print('          :No')
return p1, p2



import os

import pandas as pdimport numpy as nppath=r'../Data/itcmatrixs/36000/'result = np.zeros((50, 50))for i in os.walk(path):

#print(i)
#print(type(i))
for m in range(50):
    for n in range(50):
        r = None
        for j in range(26):
            datapath = path+i[2][j]
            p1,p2 = ContructionAgent(datapath,m,n)
            r = int(p1>=p2)
        result[m,n] = r;

r.save('result.npy') pass
ノート冊
ソースコードとデータセットが必要な場合は[email protected]あ、ご注目ありがとうございます.