mirror of
https://github.com/gryf/coach.git
synced 2025-12-17 19:20:19 +01:00
74db141d5e
* SAC algorithm * SAC - updates to agent (learn_from_batch), sac_head and sac_q_head to fix problem in gradient calculation. Now SAC agents is able to train. gym_environment - fixing an error in access to gym.spaces * Soft Actor Critic - code cleanup * code cleanup * V-head initialization fix * SAC benchmarks * SAC Documentation * typo fix * documentation fixes * documentation and version update * README typo
337 lines
15 KiB
HTML
337 lines
15 KiB
HTML
|
||
|
||
<!DOCTYPE html>
|
||
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
|
||
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
|
||
<head>
|
||
<meta charset="utf-8">
|
||
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||
|
||
<title>Persistent Advantage Learning — Reinforcement Learning Coach 0.11.0 documentation</title>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
|
||
<link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
|
||
<link rel="stylesheet" href="../../../_static/css/custom.css" type="text/css" />
|
||
<link rel="index" title="Index" href="../../../genindex.html" />
|
||
<link rel="search" title="Search" href="../../../search.html" />
|
||
<link rel="next" title="Policy Gradient" href="../policy_optimization/pg.html" />
|
||
<link rel="prev" title="Neural Episodic Control" href="nec.html" />
|
||
<link href="../../../_static/css/custom.css" rel="stylesheet" type="text/css">
|
||
|
||
|
||
|
||
<script src="../../../_static/js/modernizr.min.js"></script>
|
||
|
||
</head>
|
||
|
||
<body class="wy-body-for-nav">
|
||
|
||
|
||
<div class="wy-grid-for-nav">
|
||
|
||
|
||
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
||
<div class="wy-side-scroll">
|
||
<div class="wy-side-nav-search">
|
||
|
||
|
||
|
||
<a href="../../../index.html" class="icon icon-home"> Reinforcement Learning Coach
|
||
|
||
|
||
|
||
|
||
<img src="../../../_static/dark_logo.png" class="logo" alt="Logo"/>
|
||
|
||
</a>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<div role="search">
|
||
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
||
<input type="text" name="q" placeholder="Search docs" />
|
||
<input type="hidden" name="check_keywords" value="yes" />
|
||
<input type="hidden" name="area" value="default" />
|
||
</form>
|
||
</div>
|
||
|
||
|
||
</div>
|
||
|
||
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||
<ul>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../usage.html">Usage</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../dist_usage.html">Usage - Distributed Coach</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../features/index.html">Features</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../dashboard.html">Coach Dashboard</a></li>
|
||
</ul>
|
||
<p class="caption"><span class="caption-text">Design</span></p>
|
||
<ul>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../design/control_flow.html">Control Flow</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../design/network.html">Network Design</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../design/horizontal_scaling.html">Distributed Coach - Horizontal Scale-Out</a></li>
|
||
</ul>
|
||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||
<ul>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_agent.html">Adding a New Agent</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../../contributing/add_env.html">Adding a New Environment</a></li>
|
||
</ul>
|
||
<p class="caption"><span class="caption-text">Components</span></p>
|
||
<ul class="current">
|
||
<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Agents</a><ul class="current">
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/ac.html">Actor-Critic</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/acer.html">ACER</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../imitation/bc.html">Behavioral Cloning</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="bs_dqn.html">Bootstrapped DQN</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="categorical_dqn.html">Categorical DQN</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../imitation/cil.html">Conditional Imitation Learning</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/cppo.html">Clipped Proximal Policy Optimization</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/ddpg.html">Deep Deterministic Policy Gradient</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/sac.html">Soft Actor-Critic</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../other/dfp.html">Direct Future Prediction</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="double_dqn.html">Double DQN</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="dqn.html">Deep Q Networks</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="dueling_dqn.html">Dueling DQN</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="mmc.html">Mixed Monte Carlo</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="n_step.html">N-Step Q Learning</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="naf.html">Normalized Advantage Functions</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="nec.html">Neural Episodic Control</a></li>
|
||
<li class="toctree-l2 current"><a class="current reference internal" href="#">Persistent Advantage Learning</a><ul>
|
||
<li class="toctree-l3"><a class="reference internal" href="#network-structure">Network Structure</a></li>
|
||
<li class="toctree-l3"><a class="reference internal" href="#algorithm-description">Algorithm Description</a><ul>
|
||
<li class="toctree-l4"><a class="reference internal" href="#training-the-network">Training the network</a></li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/pg.html">Policy Gradient</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../policy_optimization/ppo.html">Proximal Policy Optimization</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="rainbow.html">Rainbow</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="qr_dqn.html">Quantile Regression DQN</a></li>
|
||
</ul>
|
||
</li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../architectures/index.html">Architectures</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../data_stores/index.html">Data Stores</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../environments/index.html">Environments</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../exploration_policies/index.html">Exploration Policies</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../filters/index.html">Filters</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../memories/index.html">Memories</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../memory_backends/index.html">Memory Backends</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../orchestrators/index.html">Orchestrators</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../core_types.html">Core Types</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../spaces.html">Spaces</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../../additional_parameters.html">Additional Parameters</a></li>
|
||
</ul>
|
||
|
||
|
||
|
||
</div>
|
||
</div>
|
||
</nav>
|
||
|
||
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
|
||
|
||
|
||
<nav class="wy-nav-top" aria-label="top navigation">
|
||
|
||
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
||
<a href="../../../index.html">Reinforcement Learning Coach</a>
|
||
|
||
</nav>
|
||
|
||
|
||
<div class="wy-nav-content">
|
||
|
||
<div class="rst-content">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<div role="navigation" aria-label="breadcrumbs navigation">
|
||
|
||
<ul class="wy-breadcrumbs">
|
||
|
||
<li><a href="../../../index.html">Docs</a> »</li>
|
||
|
||
<li><a href="../index.html">Agents</a> »</li>
|
||
|
||
<li>Persistent Advantage Learning</li>
|
||
|
||
|
||
<li class="wy-breadcrumbs-aside">
|
||
|
||
|
||
<a href="../../../_sources/components/agents/value_optimization/pal.rst.txt" rel="nofollow"> View page source</a>
|
||
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
|
||
|
||
<hr/>
|
||
</div>
|
||
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
||
<div itemprop="articleBody">
|
||
|
||
<div class="section" id="persistent-advantage-learning">
|
||
<h1>Persistent Advantage Learning<a class="headerlink" href="#persistent-advantage-learning" title="Permalink to this headline">¶</a></h1>
|
||
<p><strong>Actions space:</strong> Discrete</p>
|
||
<p><strong>References:</strong> <a class="reference external" href="https://arxiv.org/abs/1512.04860">Increasing the Action Gap: New Operators for Reinforcement Learning</a></p>
|
||
<div class="section" id="network-structure">
|
||
<h2>Network Structure<a class="headerlink" href="#network-structure" title="Permalink to this headline">¶</a></h2>
|
||
<img alt="../../../_images/dqn.png" class="align-center" src="../../../_images/dqn.png" />
|
||
</div>
|
||
<div class="section" id="algorithm-description">
|
||
<h2>Algorithm Description<a class="headerlink" href="#algorithm-description" title="Permalink to this headline">¶</a></h2>
|
||
<div class="section" id="training-the-network">
|
||
<h3>Training the network<a class="headerlink" href="#training-the-network" title="Permalink to this headline">¶</a></h3>
|
||
<ol class="arabic simple">
|
||
<li>Sample a batch of transitions from the replay buffer.</li>
|
||
<li>Start by calculating the initial target values in the same manner as they are calculated in DDQN
|
||
<span class="math notranslate nohighlight">\(y_t^{DDQN}=r(s_t,a_t )+\gamma Q(s_{t+1},argmax_a Q(s_{t+1},a))\)</span></li>
|
||
<li>The action gap <span class="math notranslate nohighlight">\(V(s_t )-Q(s_t,a_t)\)</span> should then be subtracted from each of the calculated targets.
|
||
To calculate the action gap, run the target network using the current states and get the <span class="math notranslate nohighlight">\(Q\)</span> values
|
||
for all the actions. Then estimate <span class="math notranslate nohighlight">\(V\)</span> as the maximum predicted <span class="math notranslate nohighlight">\(Q\)</span> value for the current state:
|
||
<span class="math notranslate nohighlight">\(V(s_t )=max_a Q(s_t,a)\)</span></li>
|
||
<li>For <em>advantage learning (AL)</em>, reduce the action gap weighted by a predefined parameter <span class="math notranslate nohighlight">\(\alpha\)</span> from
|
||
the targets <span class="math notranslate nohighlight">\(y_t^{DDQN}\)</span>:
|
||
<span class="math notranslate nohighlight">\(y_t=y_t^{DDQN}-\alpha \cdot (V(s_t )-Q(s_t,a_t ))\)</span></li>
|
||
<li>For <em>persistent advantage learning (PAL)</em>, the target network is also used in order to calculate the action
|
||
gap for the next state:
|
||
<span class="math notranslate nohighlight">\(V(s_{t+1} )-Q(s_{t+1},a_{t+1})\)</span>
|
||
where <span class="math notranslate nohighlight">\(a_{t+1}\)</span> is chosen by running the next states through the online network and choosing the action that
|
||
has the highest predicted <span class="math notranslate nohighlight">\(Q\)</span> value. Finally, the targets will be defined as -
|
||
<span class="math notranslate nohighlight">\(y_t=y_t^{DDQN}-\alpha \cdot min(V(s_t )-Q(s_t,a_t ),V(s_{t+1} )-Q(s_{t+1},a_{t+1} ))\)</span></li>
|
||
<li>Train the online network using the current states as inputs, and with the aforementioned targets.</li>
|
||
<li>Once in every few thousand steps, copy the weights from the online network to the target network.</li>
|
||
</ol>
|
||
<dl class="class">
|
||
<dt id="rl_coach.agents.pal_agent.PALAlgorithmParameters">
|
||
<em class="property">class </em><code class="descclassname">rl_coach.agents.pal_agent.</code><code class="descname">PALAlgorithmParameters</code><a class="reference internal" href="../../../_modules/rl_coach/agents/pal_agent.html#PALAlgorithmParameters"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#rl_coach.agents.pal_agent.PALAlgorithmParameters" title="Permalink to this definition">¶</a></dt>
|
||
<dd><table class="docutils field-list" frame="void" rules="none">
|
||
<col class="field-name" />
|
||
<col class="field-body" />
|
||
<tbody valign="top">
|
||
<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple">
|
||
<li><strong>pal_alpha</strong> – (float)
|
||
A factor that weights the amount by which the advantage learning update will be taken into account.</li>
|
||
<li><strong>persistent_advantage_learning</strong> – (bool)
|
||
If set to True, the persistent mode of advantage learning will be used, which encourages the agent to take
|
||
the same actions one after the other instead of changing actions.</li>
|
||
<li><strong>monte_carlo_mixing_rate</strong> – (float)
|
||
The amount of monte carlo values to mix into the targets of the network. The monte carlo values are just the
|
||
total discounted returns, and they can help reduce the time it takes for the network to update to the newly
|
||
seen values, since it is not based on bootstrapping the current network values.</li>
|
||
</ul>
|
||
</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</dd></dl>
|
||
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
</div>
|
||
|
||
</div>
|
||
<footer>
|
||
|
||
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
|
||
|
||
<a href="../policy_optimization/pg.html" class="btn btn-neutral float-right" title="Policy Gradient" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
|
||
|
||
|
||
<a href="nec.html" class="btn btn-neutral" title="Neural Episodic Control" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
|
||
|
||
</div>
|
||
|
||
|
||
<hr/>
|
||
|
||
<div role="contentinfo">
|
||
<p>
|
||
© Copyright 2018, Intel AI Lab
|
||
|
||
</p>
|
||
</div>
|
||
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
||
|
||
</footer>
|
||
|
||
</div>
|
||
</div>
|
||
|
||
</section>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
|
||
<script type="text/javascript" src="../../../_static/jquery.js"></script>
|
||
<script type="text/javascript" src="../../../_static/underscore.js"></script>
|
||
<script type="text/javascript" src="../../../_static/doctools.js"></script>
|
||
<script type="text/javascript" src="../../../_static/language_data.js"></script>
|
||
<script async="async" type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/latest.js?config=TeX-AMS-MML_HTMLorMML"></script>
|
||
|
||
|
||
|
||
|
||
<script type="text/javascript" src="../../../_static/js/theme.js"></script>
|
||
|
||
<script type="text/javascript">
|
||
jQuery(function () {
|
||
SphinxRtdTheme.Navigation.enable(true);
|
||
});
|
||
</script>
|
||
|
||
</body>
|
||
</html> |