A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit <a rel="external noopener" href="https://aaai.org/ojs/index.php/AAAI/article/download/6021/5877">the original URL</a>. The file type is <code>application/pdf</code>.
Adaptive Trust Region Policy Optimization: Global Convergence and Faster Rates for Regularized MDPs
<span title="2020-04-03">2020</span>
<i title="Association for the Advancement of Artificial Intelligence (AAAI)">
<a target="_blank" rel="noopener" href="https://fatcat.wiki/container/wtjcymhabjantmdtuptkk62mlq" style="color: black;">PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE</a>
</i>
Preserved Fulltext
Trust region policy optimization (TRPO) is a popular and empirically successful policy search algorithm in Reinforcement Learning (RL) in which a surrogate problem, that restricts consecutive policies to be 'close' to one another, is iteratively solved. Nevertheless, TRPO has been considered a heuristic algorithm inspired by Conservative Policy Iteration (CPI). We show that the adaptive scaling mechanism used in TRPO is in fact the natural "RL version" of traditional trust-region methods from
<span class="external-identifiers">
<a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v34i04.6021">doi:10.1609/aaai.v34i04.6021</a>
<a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4b6bmrimubejze4a6abanoz2f4">fatcat:4b6bmrimubejze4a6abanoz2f4</a>
</span>
more »
... nvex analysis. We first analyze TRPO in the planning setting, in which we have access to the model and the entire state space. Then, we consider sample-based TRPO and establish Õ(1/√N) convergence rate to the global optimum. Importantly, the adaptive scaling mechanism allows us to analyze TRPO in regularized MDPs for which we prove fast rates of Õ(1/N), much like results in convex optimization. This is the first result in RL of better rates when regularizing the instantaneous cost or reward.
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201103215601/https://aaai.org/ojs/index.php/AAAI/article/download/6021/5877" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext">
<button class="ui simple right pointing dropdown compact black labeled icon button serp-button">
<i class="icon ia-icon"></i>
Web Archive
[PDF]
<div class="menu fulltext-thumbnail">
<img src="https://blobs.fatcat.wiki/thumbnail/pdf/3d/af/3daf38ec85a848b733320598925314868c0eaa19.180px.jpg" alt="fulltext thumbnail" loading="lazy">
</div>
</button>
</a>
<a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v34i04.6021">
<button class="ui left aligned compact blue labeled icon button serp-button">
<i class="external alternate icon"></i>
Publisher / doi.org
</button>
</a>