-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtese_spy.aux
251 lines (251 loc) · 14 KB
/
tese_spy.aux
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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
\relax
\providecommand\hyper@newdestlabel[2]{}
\providecommand\HyperFirstAtBeginDocument{\AtBeginDocument}
\HyperFirstAtBeginDocument{\ifx\hyper@anchor\@undefined
\global\let\oldcontentsline\contentsline
\gdef\contentsline#1#2#3#4{\oldcontentsline{#1}{#2}{#3}}
\global\let\oldnewlabel\newlabel
\gdef\newlabel#1#2{\newlabelxx{#1}#2}
\gdef\newlabelxx#1#2#3#4#5#6{\oldnewlabel{#1}{{#2}{#3}}}
\AtEndDocument{\ifx\hyper@anchor\@undefined
\let\contentsline\oldcontentsline
\let\newlabel\oldnewlabel
\fi}
\fi}
\global\let\hyper@last\relax
\gdef\HyperFirstAtBeginDocument#1{#1}
\providecommand\HyField@AuxAddToFields[1]{}
\providecommand\HyField@AuxAddToCoFields[2]{}
\citation{abnt-url-package=hyperref}
\citation{abnt-emphasize=bf}
\citation{abnt-etal-cite=2}
\citation{abnt-etal-list=0}
\citation{abnt-etal-text=it}
\select@language{english}
\@writefile{toc}{\select@language{english}}
\@writefile{lof}{\select@language{english}}
\@writefile{lot}{\select@language{english}}
\gdef \LT@i {\LT@entry
{2}{59.75963pt}\LT@entry
{1}{270.28549pt}}
\gdef \LT@ii {\LT@entry
{1}{154.24649pt}\LT@entry
{2}{298.3846pt}}
\citation{AIapplications}
\citation{RLNature2015}
\citation{AlphaGoZero}
\@writefile{toc}{\contentsline {chapter}{\numberline {1}Introduction}{14}{chapter.1}}
\@writefile{lof}{\addvspace {10\p@ }}
\@writefile{lot}{\addvspace {10\p@ }}
\@writefile{loa}{\addvspace {10\p@ }}
\newlabel{chap:introduction}{{1}{14}{Introduction}{chapter.1}{}}
\@writefile{toc}{\contentsline {section}{\numberline {1.1}Motivation}{14}{section.1.1}}
\@writefile{lof}{\contentsline {figure}{\numberline {1.1}{\ignorespaces DeepMind recent achievements.\relax }}{15}{figure.caption.8}}
\providecommand*\caption@xref[2]{\@setref\relax\@undefined{#1}}
\newlabel{fig:deepmind_examples}{{1.1}{15}{DeepMind recent achievements.\relax }{figure.caption.8}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {1.2}{\ignorespaces Simulated humanoid agent movements and AI.\relax }}{16}{figure.caption.9}}
\newlabel{fig:humanoid_AI}{{1.2}{16}{Simulated humanoid agent movements and AI.\relax }{figure.caption.9}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {1.3}{\ignorespaces Robocup symbol and Soccer 3D Simulation league match.\relax }}{16}{figure.caption.10}}
\newlabel{fig:robocup}{{1.3}{16}{Robocup symbol and Soccer 3D Simulation league match.\relax }{figure.caption.10}{}}
\@writefile{toc}{\contentsline {section}{\numberline {1.2}Problem Statement}{16}{section.1.2}}
\citation{deepmimic}
\citation{TDLearning}
\citation{QLearning}
\citation{REINFORCE}
\citation{DDPG}
\citation{TRPO}
\citation{PPO}
\citation{RLNature2015}
\citation{ReplayBuffer}
\citation{RLNature2015}
\@writefile{toc}{\contentsline {section}{\numberline {1.3}Approach}{17}{section.1.3}}
\@writefile{toc}{\contentsline {section}{\numberline {1.4}Literature Review}{17}{section.1.4}}
\citation{deepmind1}
\citation{BengioCurrLearning}
\citation{deepmind2}
\citation{deepmind3}
\citation{gail}
\citation{deepmimic}
\citation{abbas}
\citation{abbas}
\citation{deepmimic}
\citation{TGMuzio}
\@writefile{toc}{\contentsline {section}{\numberline {1.5}Contributions}{19}{section.1.5}}
\@writefile{toc}{\contentsline {section}{\numberline {1.6}Outline of this Dissertation}{19}{section.1.6}}
\citation{Sutton1998}
\@writefile{toc}{\contentsline {chapter}{\numberline {2}Reinforcement Learning}{20}{chapter.2}}
\@writefile{lof}{\addvspace {10\p@ }}
\@writefile{lot}{\addvspace {10\p@ }}
\@writefile{loa}{\addvspace {10\p@ }}
\newlabel{chap:rl}{{2}{20}{Reinforcement Learning}{chapter.2}{}}
\@writefile{toc}{\contentsline {section}{\numberline {2.1}Model Introduction}{20}{section.2.1}}
\@writefile{lof}{\contentsline {figure}{\numberline {2.1}{\ignorespaces Agent interacting with environment.\relax }}{20}{figure.caption.11}}
\newlabel{fig:RL_basic_model}{{2.1}{20}{Agent interacting with environment.\relax }{figure.caption.11}{}}
\@writefile{toc}{\contentsline {section}{\numberline {2.2}Markov Decision Processes}{21}{section.2.2}}
\newlabel{eq:markov_basic}{{2.1}{21}{Markov Decision Processes}{equation.2.2.1}{}}
\newlabel{eq:state_value_function_definition}{{2.4}{22}{Markov Decision Processes}{equation.2.2.4}{}}
\newlabel{eq:action_value_function_definition}{{2.5}{22}{Markov Decision Processes}{equation.2.2.5}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {2.2.1}Optimality in Reinforcement Learning}{22}{subsection.2.2.1}}
\@writefile{toc}{\contentsline {subsubsection}{\numberline {2.2.1.1}Optimal Value Function}{22}{subsubsection.2.2.1.1}}
\@writefile{toc}{\contentsline {subsubsection}{\numberline {2.2.1.2}Optimal Policy}{22}{subsubsection.2.2.1.2}}
\newlabel{eq:policy_ordering}{{2.6}{22}{Optimal Policy}{equation.2.2.6}{}}
\citation{lecture1DS}
\newlabel{eq:policy_from_greedy}{{2.7}{23}{Optimal Policy}{equation.2.2.7}{}}
\@writefile{toc}{\contentsline {section}{\numberline {2.3}RL Algorithms}{23}{section.2.3}}
\@writefile{toc}{\contentsline {subsection}{\numberline {2.3.1}Categorizing RL}{23}{subsection.2.3.1}}
\citation{Sutton1998}
\@writefile{lof}{\contentsline {figure}{\numberline {2.2}{\ignorespaces Classical division among RL algorithms classifications.\relax }}{24}{figure.caption.12}}
\newlabel{fig:categorizing_RL}{{2.2}{24}{Classical division among RL algorithms classifications.\relax }{figure.caption.12}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {2.3.2}Value Function Methods}{24}{subsection.2.3.2}}
\@writefile{toc}{\contentsline {subsubsection}{\numberline {2.3.2.1}Monte Carlo Methods}{24}{subsubsection.2.3.2.1}}
\citation{Sutton1998}
\citation{TDLearning}
\newlabel{eq:epsilon-greedy}{{2.8}{25}{Monte Carlo Methods}{equation.2.3.8}{}}
\@writefile{toc}{\contentsline {subsubsection}{\numberline {2.3.2.2}Temporal-Difference and Sarsa}{25}{subsubsection.2.3.2.2}}
\newlabel{eq:TD0_update}{{2.9}{25}{Temporal-Difference and Sarsa}{equation.2.3.9}{}}
\citation{Sutton1998}
\citation{QLearning}
\@writefile{loa}{\contentsline {algocf}{\numberline {1}{\ignorespaces Sarsa algorithm\relax }}{26}{algocf.1}}
\newlabel{algo:sarsa}{{1}{26}{Temporal-Difference and Sarsa}{algocf.1}{}}
\newlabel{eq:sarsa_lambda}{{2.11}{26}{Temporal-Difference and Sarsa}{equation.2.3.11}{}}
\@writefile{toc}{\contentsline {subsubsection}{\numberline {2.3.2.3}Q-Learning}{26}{subsubsection.2.3.2.3}}
\newlabel{eq:importance_sampling}{{2.12}{26}{Q-Learning}{equation.2.3.12}{}}
\newlabel{eq:qlearning_update}{{2.13}{27}{Q-Learning}{equation.2.3.13}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {2.3.3}Policy Search Methods}{27}{subsection.2.3.3}}
\newlabel{eq:tau_def}{{2.14}{27}{Policy Search Methods}{equation.2.3.14}{}}
\newlabel{eq:theta_max}{{2.15}{27}{Policy Search Methods}{equation.2.3.15}{}}
\newlabel{eq:rewrite_objective}{{2.16}{27}{Policy Search Methods}{equation.2.3.16}{}}
\citation{REINFORCE}
\newlabel{eq:grad_with_scorefunc}{{2.17}{28}{Policy Search Methods}{equation.2.3.17}{}}
\@writefile{loa}{\contentsline {algocf}{\numberline {2}{\ignorespaces REINFORCE algorithm\relax }}{28}{algocf.2}}
\newlabel{algo:reinforce}{{2}{28}{Policy Search Methods}{algocf.2}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {2.3.4}Actor-Critic}{28}{subsection.2.3.4}}
\newlabel{eq:actor_critic_gradient}{{2.18}{28}{Actor-Critic}{equation.2.3.18}{}}
\@writefile{toc}{\contentsline {chapter}{\numberline {3}Deep Reinforcement Learning}{29}{chapter.3}}
\@writefile{lof}{\addvspace {10\p@ }}
\@writefile{lot}{\addvspace {10\p@ }}
\@writefile{loa}{\addvspace {10\p@ }}
\newlabel{chap:deep_rl}{{3}{29}{Deep Reinforcement Learning}{chapter.3}{}}
\@writefile{toc}{\contentsline {section}{\numberline {3.1}Neural Networks}{29}{section.3.1}}
\@writefile{toc}{\contentsline {subsection}{\numberline {3.1.1}Representation}{29}{subsection.3.1.1}}
\newlabel{eq:common_activation_functions}{{3.1}{30}{Representation}{equation.3.1.1}{}}
\newlabel{eq:linear_update_nn}{{3.2}{30}{Representation}{equation.3.1.2}{}}
\newlabel{eq:non_linear_update_nn}{{3.3}{30}{Representation}{equation.3.1.3}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {3.1}{\ignorespaces Shallow Neural Network Architecture.\relax }}{30}{figure.caption.13}}
\newlabel{fig:nn_basic_architecture}{{3.1}{30}{Shallow Neural Network Architecture.\relax }{figure.caption.13}{}}
\newlabel{eq:quadratic_cost}{{3.4}{31}{Representation}{equation.3.1.4}{}}
\newlabel{eq:cross_entropy_cost}{{3.5}{31}{Representation}{equation.3.1.5}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {3.1.2}Vectorization}{31}{subsection.3.1.2}}
\newlabel{eq:vectorization_concatenation_X}{{3.6}{31}{Vectorization}{equation.3.1.6}{}}
\newlabel{eq:vectorization_concatenation}{{3.7}{31}{Vectorization}{equation.3.1.7}{}}
\newlabel{eq:linear_update_nn_vectorized}{{3.8}{32}{Vectorization}{equation.3.1.8}{}}
\newlabel{eq:non_linear_update_nn_vectorized}{{3.9}{32}{Vectorization}{equation.3.1.9}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {3.1.3}Forward Propagation}{32}{subsection.3.1.3}}
\@writefile{loa}{\contentsline {algocf}{\numberline {3}{\ignorespaces Forward Propagation\relax }}{32}{algocf.3}}
\newlabel{algo:forward_propagation}{{3}{32}{Forward Propagation}{algocf.3}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {3.1.4}Backward Propagation}{32}{subsection.3.1.4}}
\newlabel{eq:partial_derivatives_notation_first}{{3.10}{32}{Backward Propagation}{equation.3.1.10}{}}
\newlabel{eq:partial_derivatives_notation_final}{{3.13}{32}{Backward Propagation}{equation.3.1.13}{}}
\citation{TRPO}
\@writefile{loa}{\contentsline {algocf}{\numberline {4}{\ignorespaces Backward Propagation\relax }}{33}{algocf.4}}
\newlabel{algo:backward_propagation}{{4}{33}{Backward Propagation}{algocf.4}{}}
\@writefile{loa}{\contentsline {algocf}{\numberline {5}{\ignorespaces Gradient Descent\relax }}{33}{algocf.5}}
\newlabel{algo:gradient_descent}{{5}{33}{Backward Propagation}{algocf.5}{}}
\@writefile{toc}{\contentsline {section}{\numberline {3.2}Trust Region Policy Optimization}{33}{section.3.2}}
\newlabel{eq:KL_divergence}{{3.14}{34}{Trust Region Policy Optimization}{equation.3.2.14}{}}
\newlabel{eq:TRPO_1}{{3.15}{34}{Trust Region Policy Optimization}{equation.3.2.15}{}}
\newlabel{eq:TRPO_2}{{3.16}{34}{Trust Region Policy Optimization}{equation.3.2.16}{}}
\citation{PPO}
\citation{PPO}
\@writefile{loa}{\contentsline {algocf}{\numberline {6}{\ignorespaces TRPO\relax }}{35}{algocf.6}}
\newlabel{algo:TRPO}{{6}{35}{Trust Region Policy Optimization}{algocf.6}{}}
\@writefile{toc}{\contentsline {section}{\numberline {3.3}Proximal Policy Optimization (PPO)}{35}{section.3.3}}
\newlabel{eq:def_r_t}{{3.17}{35}{Proximal Policy Optimization (PPO)}{equation.3.3.17}{}}
\newlabel{eq:PPO_objective_function}{{3.18}{35}{Proximal Policy Optimization (PPO)}{equation.3.3.18}{}}
\bibdata{abnt-options,Referencias/referencias}
\newlabel{eq:clip_function}{{3.19}{36}{Proximal Policy Optimization (PPO)}{equation.3.3.19}{}}
\bibcite{abbas}{Abdolmaleki \textit {et al.} 2016}
\bibciteEXPL{abbas}{Abdolmaleki \textit {et al.}}
\bibciteIMPL{abbas}{ABDOLMALEKI \textit {et al.}}
\bibciteYEAR{abbas}{2016}
\bibcite{TDLearning}{Barto \textit {et al.} 1983}
\bibciteEXPL{TDLearning}{Barto \textit {et al.}}
\bibciteIMPL{TDLearning}{BARTO \textit {et al.}}
\bibciteYEAR{TDLearning}{1983}
\bibcite{BengioCurrLearning}{Bengio \textit {et al.} 2009}
\bibciteEXPL{BengioCurrLearning}{Bengio \textit {et al.}}
\bibciteIMPL{BengioCurrLearning}{BENGIO \textit {et al.}}
\bibciteYEAR{BengioCurrLearning}{2009}
\bibcite{deepmind1}{Heess \textit {et al.} 2017}
\bibciteEXPL{deepmind1}{Heess \textit {et al.}}
\bibciteIMPL{deepmind1}{HEESS \textit {et al.}}
\bibciteYEAR{deepmind1}{2017}
\bibcite{gail}{Ho e Ermon 2016}
\bibciteEXPL{gail}{Ho e Ermon}
\bibciteIMPL{gail}{HO; ERMON}
\bibciteYEAR{gail}{2016}
\bibcite{DDPG}{Lillicrap \textit {et al.} 2015}
\bibciteEXPL{DDPG}{Lillicrap \textit {et al.}}
\bibciteIMPL{DDPG}{LILLICRAP \textit {et al.}}
\bibciteYEAR{DDPG}{2015}
\bibcite{ReplayBuffer}{Lin 1992}
\bibciteEXPL{ReplayBuffer}{Lin}
\bibciteIMPL{ReplayBuffer}{LIN}
\bibciteYEAR{ReplayBuffer}{1992}
\bibcite{AIapplications}{Marr}
\bibciteEXPL{AIapplications}{Marr}
\bibciteIMPL{AIapplications}{MARR}
\bibciteYEAR{AIapplications}{}
\bibcite{deepmind2}{Merel \textit {et al.} 2017}
\bibciteEXPL{deepmind2}{Merel \textit {et al.}}
\bibciteIMPL{deepmind2}{MEREL \textit {et al.}}
\bibciteYEAR{deepmind2}{2017}
\bibcite{RLNature2015}{Mnih \textit {et al.} 2015}
\bibciteEXPL{RLNature2015}{Mnih \textit {et al.}}
\bibciteIMPL{RLNature2015}{MNIH \textit {et al.}}
\bibciteYEAR{RLNature2015}{2015}
\@writefile{toc}{\contentsline {chapter}{Bibliography}{37}{chapter.4}}
\bibcite{TGMuzio}{Muzio \textit {et al.} 2017}
\bibciteEXPL{TGMuzio}{Muzio \textit {et al.}}
\bibciteIMPL{TGMuzio}{MUZIO \textit {et al.}}
\bibciteYEAR{TGMuzio}{2017}
\bibcite{deepmimic}{Peng \textit {et al.} 2018}
\bibciteEXPL{deepmimic}{Peng \textit {et al.}}
\bibciteIMPL{deepmimic}{PENG \textit {et al.}}
\bibciteYEAR{deepmimic}{2018}
\bibcite{TRPO}{Schulman \textit {et al.} 2015}
\bibciteEXPL{TRPO}{Schulman \textit {et al.}}
\bibciteIMPL{TRPO}{SCHULMAN \textit {et al.}}
\bibciteYEAR{TRPO}{2015}
\bibcite{PPO}{Schulman \textit {et al.} 2017}
\bibciteEXPL{PPO}{Schulman \textit {et al.}}
\bibciteIMPL{PPO}{SCHULMAN \textit {et al.}}
\bibciteYEAR{PPO}{2017}
\bibcite{lecture1DS}{Silver}
\bibciteEXPL{lecture1DS}{Silver}
\bibciteIMPL{lecture1DS}{SILVER}
\bibciteYEAR{lecture1DS}{}
\bibcite{AlphaGoZero}{Silver \textit {et al.} 2017}
\bibciteEXPL{AlphaGoZero}{Silver \textit {et al.}}
\bibciteIMPL{AlphaGoZero}{SILVER \textit {et al.}}
\bibciteYEAR{AlphaGoZero}{2017}
\bibcite{Sutton1998}{Sutton e Barto 1998}
\bibciteEXPL{Sutton1998}{Sutton e Barto}
\bibciteIMPL{Sutton1998}{SUTTON; BARTO}
\bibciteYEAR{Sutton1998}{1998}
\bibcite{deepmind3}{Wang \textit {et al.} 2017}
\bibciteEXPL{deepmind3}{Wang \textit {et al.}}
\bibciteIMPL{deepmind3}{WANG \textit {et al.}}
\bibciteYEAR{deepmind3}{2017}
\bibcite{QLearning}{Watkins 1989}
\bibciteEXPL{QLearning}{Watkins}
\bibciteIMPL{QLearning}{WATKINS}
\bibciteYEAR{QLearning}{1989}
\bibcite{REINFORCE}{Wiliams 1992}
\bibciteEXPL{REINFORCE}{Wiliams}
\bibciteIMPL{REINFORCE}{WILIAMS}
\bibciteYEAR{REINFORCE}{1992}
\newlabel{LastPage}{{}{39}{}{page.39}{}}
\xdef\lastpage@lastpage{39}
\xdef\lastpage@lastpageHy{39}
\bibstyle{abnt-alf}