Efficient Wasserstein Natural Gradients for Reinforcement Learning [article]

Ted Moskovitz, Michael Arbel, Ferenc Huszar, Arthur Gretton
<span title="2021-03-18">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
A novel optimization approach is proposed for application to policy gradient methods and evolution strategies for reinforcement learning (RL). The procedure uses a computationally efficient Wasserstein natural gradient (WNG) descent that takes advantage of the geometry induced by a Wasserstein penalty to speed optimization. This method follows the recent theme in RL of including a divergence penalty in the objective to establish a trust region. Experiments on challenging tasks demonstrate
more &raquo; ... ements in both computational cost and performance over advanced baselines.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.05380v4">arXiv:2010.05380v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7bhuob2ngjfdbfzmn4q44gyxra">fatcat:7bhuob2ngjfdbfzmn4q44gyxra</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210329045527/https://arxiv.org/pdf/2010.05380v4.pdf" 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/75/6e/756e18d759584c120d17d843e1c0be5375d80316.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.05380v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>