hot update
This commit is contained in:
johnjim0816
@@ -28,6 +28,8 @@
|
||||
\@writefile{loa}{\contentsline {algorithm}{\numberline {}{\ignorespaces }}{6}{algorithm.}\protected@file@percent }
|
||||
\@writefile{toc}{\contentsline {section}{\numberline {6}SoftQ算法}{7}{section.6}\protected@file@percent }
|
||||
\@writefile{loa}{\contentsline {algorithm}{\numberline {}{\ignorespaces }}{7}{algorithm.}\protected@file@percent }
|
||||
\@writefile{toc}{\contentsline {section}{\numberline {7}SAC算法}{8}{section.7}\protected@file@percent }
|
||||
\@writefile{toc}{\contentsline {section}{\numberline {7}SAC-S算法}{8}{section.7}\protected@file@percent }
|
||||
\@writefile{loa}{\contentsline {algorithm}{\numberline {}{\ignorespaces }}{8}{algorithm.}\protected@file@percent }
|
||||
\gdef \@abspage@last{8}
|
||||
\@writefile{toc}{\contentsline {section}{\numberline {8}SAC算法}{9}{section.8}\protected@file@percent }
|
||||
\@writefile{loa}{\contentsline {algorithm}{\numberline {}{\ignorespaces }}{9}{algorithm.}\protected@file@percent }
|
||||
\gdef \@abspage@last{9}
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
This is XeTeX, Version 3.141592653-2.6-0.999993 (TeX Live 2021) (preloaded format=xelatex 2021.8.22) 22 AUG 2022 16:54
|
||||
This is XeTeX, Version 3.141592653-2.6-0.999993 (TeX Live 2021) (preloaded format=xelatex 2021.8.22) 23 AUG 2022 19:26
|
||||
entering extended mode
|
||||
restricted \write18 enabled.
|
||||
file:line:error style messages enabled.
|
||||
@@ -415,85 +415,85 @@ Package: titlesec 2019/10/16 v2.13 Sectioning titles
|
||||
) (./pseudocodes.aux)
|
||||
\openout1 = `pseudocodes.aux'.
|
||||
|
||||
LaTeX Font Info: Checking defaults for OML/cmm/m/it on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for OMS/cmsy/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for OT1/cmr/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for T1/cmr/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for TS1/cmr/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for TU/lmr/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for OMX/cmex/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for PD1/pdf/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for PU/pdf/m/n on input line 13.
|
||||
LaTeX Font Info: ... okay on input line 13.
|
||||
LaTeX Font Info: Checking defaults for OML/cmm/m/it on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for OMS/cmsy/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for OT1/cmr/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for T1/cmr/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for TS1/cmr/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for TU/lmr/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for OMX/cmex/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for PD1/pdf/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
LaTeX Font Info: Checking defaults for PU/pdf/m/n on input line 14.
|
||||
LaTeX Font Info: ... okay on input line 14.
|
||||
ABD: EverySelectfont initializing macros
|
||||
LaTeX Info: Redefining \selectfont on input line 13.
|
||||
LaTeX Info: Redefining \selectfont on input line 14.
|
||||
|
||||
Package fontspec Info: Adjusting the maths setup (use [no-math] to avoid
|
||||
(fontspec) this).
|
||||
|
||||
\symlegacymaths=\mathgroup6
|
||||
LaTeX Font Info: Overwriting symbol font `legacymaths' in version `bold'
|
||||
(Font) OT1/cmr/m/n --> OT1/cmr/bx/n on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \acute on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \grave on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \ddot on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \tilde on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \bar on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \breve on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \check on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \hat on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \dot on input line 13.
|
||||
LaTeX Font Info: Redeclaring math accent \mathring on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Gamma on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Delta on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Theta on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Lambda on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Xi on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Pi on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Sigma on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Upsilon on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Phi on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Psi on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \Omega on input line 13.
|
||||
LaTeX Font Info: Redeclaring math symbol \mathdollar on input line 13.
|
||||
LaTeX Font Info: Redeclaring symbol font `operators' on input line 13.
|
||||
(Font) OT1/cmr/m/n --> OT1/cmr/bx/n on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \acute on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \grave on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \ddot on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \tilde on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \bar on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \breve on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \check on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \hat on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \dot on input line 14.
|
||||
LaTeX Font Info: Redeclaring math accent \mathring on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Gamma on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Delta on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Theta on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Lambda on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Xi on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Pi on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Sigma on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Upsilon on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Phi on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Psi on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \Omega on input line 14.
|
||||
LaTeX Font Info: Redeclaring math symbol \mathdollar on input line 14.
|
||||
LaTeX Font Info: Redeclaring symbol font `operators' on input line 14.
|
||||
LaTeX Font Info: Encoding `OT1' has changed to `TU' for symbol font
|
||||
(Font) `operators' in the math version `normal' on input line 13.
|
||||
(Font) `operators' in the math version `normal' on input line 14.
|
||||
LaTeX Font Info: Overwriting symbol font `operators' in version `normal'
|
||||
(Font) OT1/cmr/m/n --> TU/lmr/m/n on input line 13.
|
||||
(Font) OT1/cmr/m/n --> TU/lmr/m/n on input line 14.
|
||||
LaTeX Font Info: Encoding `OT1' has changed to `TU' for symbol font
|
||||
(Font) `operators' in the math version `bold' on input line 13.
|
||||
(Font) `operators' in the math version `bold' on input line 14.
|
||||
LaTeX Font Info: Overwriting symbol font `operators' in version `bold'
|
||||
(Font) OT1/cmr/bx/n --> TU/lmr/m/n on input line 13.
|
||||
(Font) OT1/cmr/bx/n --> TU/lmr/m/n on input line 14.
|
||||
LaTeX Font Info: Overwriting symbol font `operators' in version `normal'
|
||||
(Font) TU/lmr/m/n --> TU/lmr/m/n on input line 13.
|
||||
(Font) TU/lmr/m/n --> TU/lmr/m/n on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathit' in version `normal'
|
||||
(Font) OT1/cmr/m/it --> TU/lmr/m/it on input line 13.
|
||||
(Font) OT1/cmr/m/it --> TU/lmr/m/it on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathbf' in version `normal'
|
||||
(Font) OT1/cmr/bx/n --> TU/lmr/b/n on input line 13.
|
||||
(Font) OT1/cmr/bx/n --> TU/lmr/b/n on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathsf' in version `normal'
|
||||
(Font) OT1/cmss/m/n --> TU/lmss/m/n on input line 13.
|
||||
(Font) OT1/cmss/m/n --> TU/lmss/m/n on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathtt' in version `normal'
|
||||
(Font) OT1/cmtt/m/n --> TU/lmtt/m/n on input line 13.
|
||||
(Font) OT1/cmtt/m/n --> TU/lmtt/m/n on input line 14.
|
||||
LaTeX Font Info: Overwriting symbol font `operators' in version `bold'
|
||||
(Font) TU/lmr/m/n --> TU/lmr/b/n on input line 13.
|
||||
(Font) TU/lmr/m/n --> TU/lmr/b/n on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathit' in version `bold'
|
||||
(Font) OT1/cmr/bx/it --> TU/lmr/b/it on input line 13.
|
||||
(Font) OT1/cmr/bx/it --> TU/lmr/b/it on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathsf' in version `bold'
|
||||
(Font) OT1/cmss/bx/n --> TU/lmss/b/n on input line 13.
|
||||
(Font) OT1/cmss/bx/n --> TU/lmss/b/n on input line 14.
|
||||
LaTeX Font Info: Overwriting math alphabet `\mathtt' in version `bold'
|
||||
(Font) OT1/cmtt/m/n --> TU/lmtt/b/n on input line 13.
|
||||
Package hyperref Info: Link coloring OFF on input line 13.
|
||||
(Font) OT1/cmtt/m/n --> TU/lmtt/b/n on input line 14.
|
||||
Package hyperref Info: Link coloring OFF on input line 14.
|
||||
(/usr/local/texlive/2021/texmf-dist/tex/latex/hyperref/nameref.sty
|
||||
Package: nameref 2021-04-02 v2.47 Cross-referencing by name of section
|
||||
(/usr/local/texlive/2021/texmf-dist/tex/latex/refcount/refcount.sty
|
||||
@@ -503,9 +503,9 @@ Package: gettitlestring 2019/12/15 v1.6 Cleanup title references (HO)
|
||||
)
|
||||
\c@section@level=\count313
|
||||
)
|
||||
LaTeX Info: Redefining \ref on input line 13.
|
||||
LaTeX Info: Redefining \pageref on input line 13.
|
||||
LaTeX Info: Redefining \nameref on input line 13.
|
||||
LaTeX Info: Redefining \ref on input line 14.
|
||||
LaTeX Info: Redefining \pageref on input line 14.
|
||||
LaTeX Info: Redefining \nameref on input line 14.
|
||||
(./pseudocodes.out) (./pseudocodes.out)
|
||||
\@outlinefile=\write3
|
||||
\openout3 = `pseudocodes.out'.
|
||||
@@ -515,19 +515,19 @@ LaTeX Info: Redefining \nameref on input line 13.
|
||||
\openout4 = `pseudocodes.toc'.
|
||||
|
||||
LaTeX Font Info: Font shape `TU/SongtiSCLight(0)/m/sl' in size <10.95> not available
|
||||
(Font) Font shape `TU/SongtiSCLight(0)/m/it' tried instead on input line 16.
|
||||
(Font) Font shape `TU/SongtiSCLight(0)/m/it' tried instead on input line 17.
|
||||
[1
|
||||
|
||||
]
|
||||
Package hyperref Info: bookmark level for unknown algorithm defaults to 0 on input line 21.
|
||||
Package hyperref Info: bookmark level for unknown algorithm defaults to 0 on input line 22.
|
||||
[2
|
||||
|
||||
]
|
||||
LaTeX Font Info: Trying to load font information for U+msa on input line 31.
|
||||
LaTeX Font Info: Trying to load font information for U+msa on input line 32.
|
||||
(/usr/local/texlive/2021/texmf-dist/tex/latex/amsfonts/umsa.fd
|
||||
File: umsa.fd 2013/01/14 v3.01 AMS symbols A
|
||||
)
|
||||
LaTeX Font Info: Trying to load font information for U+msb on input line 31.
|
||||
LaTeX Font Info: Trying to load font information for U+msb on input line 32.
|
||||
(/usr/local/texlive/2021/texmf-dist/tex/latex/amsfonts/umsb.fd
|
||||
File: umsb.fd 2013/01/14 v3.01 AMS symbols B
|
||||
) [3
|
||||
@@ -536,38 +536,35 @@ File: umsb.fd 2013/01/14 v3.01 AMS symbols B
|
||||
|
||||
] [5
|
||||
|
||||
]
|
||||
Underfull \hbox (badness 10000) in paragraph at lines 111--112
|
||||
[] []\TU/SongtiSCLight(0)/m/n/10.95 计 算 实 际 的 $\OML/cmm/m/it/10.95 Q$ \TU/SongtiSCLight(0)/m/n/10.95 值，| 即 $\OML/cmm/m/it/10.95 y[] \OT1/cmr/m/n/10.95 =
|
||||
[]
|
||||
|
||||
[6
|
||||
] [6
|
||||
|
||||
] [7
|
||||
|
||||
] [8
|
||||
|
||||
]
|
||||
Overfull \hbox (32.54117pt too wide) in paragraph at lines 183--183
|
||||
Overfull \hbox (32.54117pt too wide) in paragraph at lines 212--212
|
||||
[][]$[]\OML/cmm/m/it/9 J[]\OT1/cmr/m/n/9 (\OML/cmm/m/it/9 ^^R\OT1/cmr/m/n/9 ) = \OMS/cmsy/m/n/9 r[]\OML/cmm/m/it/9 Q[] [] []$|
|
||||
[]
|
||||
|
||||
|
||||
Overfull \hbox (15.41673pt too wide) in paragraph at lines 184--184
|
||||
Overfull \hbox (15.41673pt too wide) in paragraph at lines 213--213
|
||||
[][]$[]\OML/cmm/m/it/9 J[]\OT1/cmr/m/n/9 (\OML/cmm/m/it/9 ^^^\OT1/cmr/m/n/9 ) = \OMS/cmsy/m/n/9 r[]\OML/cmm/m/it/9 ^^K [] [] \OT1/cmr/m/n/9 + [] \OMS/cmsy/m/n/9 r[]\OML/cmm/m/it/9 f[] []$\TU/lmr/m/n/9 ,$[][] \OT1/cmr/m/n/9 =
|
||||
[]
|
||||
|
||||
[8
|
||||
[9
|
||||
|
||||
] (./pseudocodes.aux)
|
||||
Package rerunfilecheck Info: File `pseudocodes.out' has not changed.
|
||||
(rerunfilecheck) Checksum: 4575BA7458AA23D6E696EFFE39D05727;640.
|
||||
(rerunfilecheck) Checksum: 35B5A79A86EF3BC70F1A0B3BCBEBAA13;724.
|
||||
)
|
||||
Here is how much of TeX's memory you used:
|
||||
14813 strings out of 476919
|
||||
312635 string characters out of 5821840
|
||||
653471 words of memory out of 5000000
|
||||
34563 multiletter control sequences out of 15000+600000
|
||||
413601 words of font info for 90 fonts, out of 8000000 for 9000
|
||||
14827 strings out of 476919
|
||||
313456 string characters out of 5821840
|
||||
653576 words of memory out of 5000000
|
||||
34576 multiletter control sequences out of 15000+600000
|
||||
413609 words of font info for 91 fonts, out of 8000000 for 9000
|
||||
1348 hyphenation exceptions out of 8191
|
||||
101i,13n,104p,676b,736s stack positions out of 5000i,500n,10000p,200000b,80000s
|
||||
101i,13n,104p,676b,697s stack positions out of 5000i,500n,10000p,200000b,80000s
|
||||
|
||||
Output written on pseudocodes.pdf (8 pages).
|
||||
Output written on pseudocodes.pdf (9 pages).
|
||||
|
||||
@@ -4,4 +4,5 @@
|
||||
\BOOKMARK [1][-]{section.4}{\376\377\000P\000o\000l\000i\000c\000y\000\040\000G\000r\000a\000d\000i\000e\000n\000t\173\227\154\325}{}% 4
|
||||
\BOOKMARK [1][-]{section.5}{\376\377\000D\000Q\000N\173\227\154\325}{}% 5
|
||||
\BOOKMARK [1][-]{section.6}{\376\377\000S\000o\000f\000t\000Q\173\227\154\325}{}% 6
|
||||
\BOOKMARK [1][-]{section.7}{\376\377\000S\000A\000C\173\227\154\325}{}% 7
|
||||
\BOOKMARK [1][-]{section.7}{\376\377\000S\000A\000C\000-\000S\173\227\154\325}{}% 7
|
||||
\BOOKMARK [1][-]{section.8}{\376\377\000S\000A\000C\173\227\154\325}{}% 8
|
||||
|
||||
Binary file not shown.
Binary file not shown.
@@ -10,6 +10,7 @@
|
||||
\usepackage{titlesec}
|
||||
\usepackage{float} % 调用该包能够使用[H]
|
||||
% \pagestyle{plain} % 去除页眉，但是保留页脚编号，都去掉plain换empty
|
||||
|
||||
\begin{document}
|
||||
\tableofcontents % 目录，注意要运行两下或者vscode保存两下才能显示
|
||||
% \singlespacing
|
||||
@@ -88,7 +89,7 @@
|
||||
\clearpage
|
||||
\section{DQN算法}
|
||||
\begin{algorithm}[H] % [H]固定位置
|
||||
\floatname{algorithm}{{DQN算法}}
|
||||
\floatname{algorithm}{{DQN算法}{\hypersetup{linkcolor=white}\footnotemark}}
|
||||
\renewcommand{\thealgorithm}{} % 去掉算法标号
|
||||
\caption{}
|
||||
\renewcommand{\algorithmicrequire}{\textbf{输入:}}
|
||||
@@ -108,13 +109,17 @@
|
||||
\STATE 更新环境状态$s_{t+1} \leftarrow s_t$
|
||||
\STATE {\bfseries 更新策略：}
|
||||
\STATE 从$D$中采样一个batch的transition
|
||||
\STATE 计算实际的$Q$值，即$y_{j}= \begin{cases}r_{j} & \text {对于终止状态} s_{j+1} \\ r_{j}+\gamma \max _{a^{\prime}} Q\left(s_{j+1}, a^{\prime} ; \theta\right) & \text {对于非终止状态} s_{j+1}\end{cases}$
|
||||
\STATE 对损失 $\left(y_{j}-Q\left(s_{j}, a_{j} ; \theta\right)\right)^{2}$关于参数$\theta$做随机梯度下降
|
||||
\STATE 计算实际的$Q$值，即$y_{j}${\hypersetup{linkcolor=white}\footnotemark}
|
||||
\STATE 对损失 $L(\theta)=\left(y_{i}-Q\left(s_{i}, a_{i} ; \theta\right)\right)^{2}$关于参数$\theta$做随机梯度下降{\hypersetup{linkcolor=white}\footnotemark}
|
||||
\ENDFOR
|
||||
\STATE 每$C$个回合复制参数$\hat{Q}\leftarrow Q$(此处也可像原论文中放到小循环中改成每$C$步，但没有每$C$个回合稳定)
|
||||
\STATE 每$C$个回合复制参数$\hat{Q}\leftarrow Q${\hypersetup{linkcolor=white}\footnotemark}
|
||||
\ENDFOR
|
||||
\end{algorithmic}
|
||||
\end{algorithm}
|
||||
\footnotetext[1]{Playing Atari with Deep Reinforcement Learning}
|
||||
\footnotetext[2]{$y_{i}= \begin{cases}r_{i} & \text {对于终止状态} s_{i+1} \\ r_{i}+\gamma \max _{a^{\prime}} Q\left(s_{i+1}, a^{\prime} ; \theta\right) & \text {对于非终止状态} s_{i+1}\end{cases}$}
|
||||
\footnotetext[3]{$\theta_i \leftarrow \theta_i - \lambda \nabla_{\theta_{i}} L_{i}\left(\theta_{i}\right)$}
|
||||
\footnotetext[4]{此处也可像原论文中放到小循环中改成每$C$步，但没有每$C$个回合稳定}
|
||||
\clearpage
|
||||
|
||||
\section{SoftQ算法}
|
||||
@@ -153,13 +158,37 @@
|
||||
\footnotetext[2]{$J_{Q}(\theta)=\mathbb{E}_{\mathbf{s}_{t} \sim q_{\mathbf{s}_{t}}, \mathbf{a}_{t} \sim q_{\mathbf{a}_{t}}}\left[\frac{1}{2}\left(\hat{Q}_{\mathrm{soft}}^{\bar{\theta}}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)-Q_{\mathrm{soft}}^{\theta}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)\right)^{2}\right]$}
|
||||
\footnotetext[3]{$\begin{aligned} \Delta f^{\phi}\left(\cdot ; \mathbf{s}_{t}\right)=& \mathbb{E}_{\mathbf{a}_{t} \sim \pi^{\phi}}\left[\left.\kappa\left(\mathbf{a}_{t}, f^{\phi}\left(\cdot ; \mathbf{s}_{t}\right)\right) \nabla_{\mathbf{a}^{\prime}} Q_{\mathrm{soft}}^{\theta}\left(\mathbf{s}_{t}, \mathbf{a}^{\prime}\right)\right|_{\mathbf{a}^{\prime}=\mathbf{a}_{t}}\right.\\ &\left.+\left.\alpha \nabla_{\mathbf{a}^{\prime}} \kappa\left(\mathbf{a}^{\prime}, f^{\phi}\left(\cdot ; \mathbf{s}_{t}\right)\right)\right|_{\mathbf{a}^{\prime}=\mathbf{a}_{t}}\right] \end{aligned}$}
|
||||
\clearpage
|
||||
\section{SAC-S算法}
|
||||
\begin{algorithm}[H] % [H]固定位置
|
||||
\floatname{algorithm}{{SAC-S算法}\footnotemark[1]}
|
||||
\renewcommand{\thealgorithm}{} % 去掉算法标号
|
||||
\caption{}
|
||||
\begin{algorithmic}[1] % [1]显示步数
|
||||
\STATE 初始化参数$\psi, \bar{\psi}, \theta, \phi$
|
||||
\FOR {回合数 = $1,M$}
|
||||
\FOR {时步 = $1,t$}
|
||||
\STATE 根据$\boldsymbol{a}_{t} \sim \pi_{\phi}\left(\boldsymbol{a}_{t} \mid \mathbf{s}_{t}\right)$采样动作$a_t$
|
||||
\STATE 环境反馈奖励和下一个状态，$\mathbf{s}_{t+1} \sim p\left(\mathbf{s}_{t+1} \mid \mathbf{s}_{t}, \mathbf{a}_{t}\right)$
|
||||
\STATE 存储transition到经验回放中，$\mathcal{D} \leftarrow \mathcal{D} \cup\left\{\left(\mathbf{s}_{t}, \mathbf{a}_{t}, r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right), \mathbf{s}_{t+1}\right)\right\}$
|
||||
\STATE 更新环境状态$s_{t+1} \leftarrow s_t$
|
||||
\STATE {\bfseries 更新策略：}
|
||||
\STATE $\psi \leftarrow \psi-\lambda_{V} \hat{\nabla}_{\psi} J_{V}(\psi)$
|
||||
\STATE $\theta_{i} \leftarrow \theta_{i}-\lambda_{Q} \hat{\nabla}_{\theta_{i}} J_{Q}\left(\theta_{i}\right)$ for $i \in\{1,2\}$
|
||||
\STATE $\phi \leftarrow \phi-\lambda_{\pi} \hat{\nabla}_{\phi} J_{\pi}(\phi)$
|
||||
\STATE $\bar{\psi} \leftarrow \tau \psi+(1-\tau) \bar{\psi}$
|
||||
\ENDFOR
|
||||
\ENDFOR
|
||||
\end{algorithmic}
|
||||
\end{algorithm}
|
||||
\footnotetext[1]{Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor}
|
||||
\clearpage
|
||||
\section{SAC算法}
|
||||
\begin{algorithm}[H] % [H]固定位置
|
||||
\floatname{algorithm}{{Soft Actor Critic算法}}
|
||||
\floatname{algorithm}{{SAC算法}\footnotemark[1]}
|
||||
\renewcommand{\thealgorithm}{} % 去掉算法标号
|
||||
\caption{}
|
||||
\begin{algorithmic}[1]
|
||||
\STATE 初始化两个Actor的网络参数$\theta_1,\theta_2$以及一个Critic网络参数$\phi$ % 初始化
|
||||
\STATE 初始化网络参数$\theta_1,\theta_2$以及$\phi$ % 初始化
|
||||
\STATE 复制参数到目标网络$\bar{\theta_1} \leftarrow \theta_1,\bar{\theta_2} \leftarrow \theta_2,$
|
||||
\STATE 初始化经验回放$D$
|
||||
\FOR {回合数 = $1,M$}
|
||||
@@ -170,18 +199,18 @@
|
||||
\STATE 存储transition到经验回放中，$\mathcal{D} \leftarrow \mathcal{D} \cup\left\{\left(\mathbf{s}_{t}, \mathbf{a}_{t}, r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right), \mathbf{s}_{t+1}\right)\right\}$
|
||||
\STATE 更新环境状态$s_{t+1} \leftarrow s_t$
|
||||
\STATE {\bfseries 更新策略：}
|
||||
\STATE 更新$Q$函数，$\theta_{i} \leftarrow \theta_{i}-\lambda_{Q} \hat{\nabla}_{\theta_{i}} J_{Q}\left(\theta_{i}\right)$ for $i \in\{1,2\}$\footnotemark[1]\footnotemark[2]
|
||||
\STATE 更新策略权重，$\phi \leftarrow \phi-\lambda_{\pi} \hat{\nabla}_{\phi} J_{\pi}(\phi)$ \footnotemark[3]
|
||||
\STATE 调整temperature，$\alpha \leftarrow \alpha-\lambda \hat{\nabla}_{\alpha} J(\alpha)$ \footnotemark[4]
|
||||
\STATE 更新$Q$函数，$\theta_{i} \leftarrow \theta_{i}-\lambda_{Q} \hat{\nabla}_{\theta_{i}} J_{Q}\left(\theta_{i}\right)$ for $i \in\{1,2\}$\footnotemark[2]\footnotemark[3]
|
||||
\STATE 更新策略权重，$\phi \leftarrow \phi-\lambda_{\pi} \hat{\nabla}_{\phi} J_{\pi}(\phi)$ \footnotemark[4]
|
||||
\STATE 调整temperature，$\alpha \leftarrow \alpha-\lambda \hat{\nabla}_{\alpha} J(\alpha)$ \footnotemark[5]
|
||||
\STATE 更新目标网络权重，$\bar{\theta}_{i} \leftarrow \tau \theta_{i}+(1-\tau) \bar{\theta}_{i}$ for $i \in\{1,2\}$
|
||||
\ENDFOR
|
||||
\ENDFOR
|
||||
\end{algorithmic}
|
||||
|
||||
\end{algorithmic}
|
||||
\end{algorithm}
|
||||
\footnotetext[1]{$J_{Q}(\theta)=\mathbb{E}_{\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right) \sim \mathcal{D}}\left[\frac{1}{2}\left(Q_{\theta}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)-\left(r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)+\gamma \mathbb{E}_{\mathbf{s}_{t+1} \sim p}\left[V_{\bar{\theta}}\left(\mathbf{s}_{t+1}\right)\right]\right)\right)^{2}\right]$}
|
||||
\footnotetext[2]{$\hat{\nabla}_{\theta} J_{Q}(\theta)=\nabla_{\theta} Q_{\theta}\left(\mathbf{a}_{t}, \mathbf{s}_{t}\right)\left(Q_{\theta}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)-\left(r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)+\gamma\left(Q_{\bar{\theta}}\left(\mathbf{s}_{t+1}, \mathbf{a}_{t+1}\right)-\alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t+1} \mid \mathbf{s}_{t+1}\right)\right)\right)\right)\right.$}
|
||||
\footnotetext[3]{$\hat{\nabla}_{\phi} J_{\pi}(\phi)=\nabla_{\phi} \alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)\right)+\left(\nabla_{\mathbf{a}_{t}} \alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)\right)-\nabla_{\mathbf{a}_{t}} Q\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)\right) \nabla_{\phi} f_{\phi}\left(\epsilon_{t} ; \mathbf{s}_{t}\right)$,$\mathbf{a}_{t}=f_{\phi}\left(\epsilon_{t} ; \mathbf{s}_{t}\right)$}
|
||||
\footnotetext[4]{$J(\alpha)=\mathbb{E}_{\mathbf{a}_{t} \sim \pi_{t}}\left[-\alpha \log \pi_{t}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)-\alpha \overline{\mathcal{H}}\right]$}
|
||||
\footnotetext[2]{Soft Actor-Critic Algorithms and Applications}
|
||||
\footnotetext[2]{$J_{Q}(\theta)=\mathbb{E}_{\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right) \sim \mathcal{D}}\left[\frac{1}{2}\left(Q_{\theta}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)-\left(r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)+\gamma \mathbb{E}_{\mathbf{s}_{t+1} \sim p}\left[V_{\bar{\theta}}\left(\mathbf{s}_{t+1}\right)\right]\right)\right)^{2}\right]$}
|
||||
\footnotetext[3]{$\hat{\nabla}_{\theta} J_{Q}(\theta)=\nabla_{\theta} Q_{\theta}\left(\mathbf{a}_{t}, \mathbf{s}_{t}\right)\left(Q_{\theta}\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)-\left(r\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)+\gamma\left(Q_{\bar{\theta}}\left(\mathbf{s}_{t+1}, \mathbf{a}_{t+1}\right)-\alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t+1} \mid \mathbf{s}_{t+1}\right)\right)\right)\right)\right.$}
|
||||
\footnotetext[4]{$\hat{\nabla}_{\phi} J_{\pi}(\phi)=\nabla_{\phi} \alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)\right)+\left(\nabla_{\mathbf{a}_{t}} \alpha \log \left(\pi_{\phi}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)\right)-\nabla_{\mathbf{a}_{t}} Q\left(\mathbf{s}_{t}, \mathbf{a}_{t}\right)\right) \nabla_{\phi} f_{\phi}\left(\epsilon_{t} ; \mathbf{s}_{t}\right)$,$\mathbf{a}_{t}=f_{\phi}\left(\epsilon_{t} ; \mathbf{s}_{t}\right)$}
|
||||
\footnotetext[5]{$J(\alpha)=\mathbb{E}_{\mathbf{a}_{t} \sim \pi_{t}}\left[-\alpha \log \pi_{t}\left(\mathbf{a}_{t} \mid \mathbf{s}_{t}\right)-\alpha \overline{\mathcal{H}}\right]$}
|
||||
\clearpage
|
||||
\end{document}
|
||||
@@ -4,4 +4,5 @@
|
||||
\contentsline {section}{\numberline {4}Policy Gradient算法}{5}{section.4}%
|
||||
\contentsline {section}{\numberline {5}DQN算法}{6}{section.5}%
|
||||
\contentsline {section}{\numberline {6}SoftQ算法}{7}{section.6}%
|
||||
\contentsline {section}{\numberline {7}SAC算法}{8}{section.7}%
|
||||
\contentsline {section}{\numberline {7}SAC-S算法}{8}{section.7}%
|
||||
\contentsline {section}{\numberline {8}SAC算法}{9}{section.8}%
|
||||
|
||||
Reference in New Issue
Block a user