Skip to content

Commit

Permalink
Update website
Browse files Browse the repository at this point in the history
  • Loading branch information
maxencefaldor committed May 31, 2024
1 parent 5f560f0 commit 234d04b
Showing 1 changed file with 4 additions and 6 deletions.
10 changes: 4 additions & 6 deletions docs/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title"><span style="color: #e1144b;">Quality</span>-<span style="color: #0053d6;">Diversity</span> Actor-Critic:</br><span style="color: #7a7a7a;">Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics</span></h1>
<h1 class="title is-1 publication-title"><span style="color: #e1144b;">Quality</span>-<span style="color: #0053d6;">Diversity</span> Actor-Critic:<br/><span style="color: #7a7a7a;">Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics</span></h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="https://luca.grillotti.com">Luca Grillotti</a><sup>*1</sup>,</span>
Expand All @@ -50,11 +50,9 @@ <h1 class="title is-1 publication-title"><span style="color: #e1144b;">Quality</
<span class="author-block"><sup>2</sup>Iconic AI</span>
</div>

<!-- ICML 2024 -->
<div class="is-size-5 has-text-centered">
<span class="tag is-success is-light">ICML 2024</span>
</div>
<br>

<div class="column has-text-centered">
<div class="publication-links">
Expand Down Expand Up @@ -198,17 +196,17 @@ <h2 class="title is-3">Abstract</h2>
<h2 class="title is-3">Method</h2>
<p>
We formalize Quality-Diversity optimization as a constraint optimization problem. We intend to learn a skill-conditioned policy that (1) maximizes the expected return, and (2) is subject to the expected features converge to the desired skill.
</p></br>
</p><br/>
<p>
We introduce an actor-critic method that leverages two critics: a performance critic (i.e., a value function) to optimize (1), and a behavior critic (i.e., a successor features) to optimize (2).
</p></br>
</p><br/>
<p>
The actor optimizes an objective that seamlessly unifies both critics using constrained optimization to (1) maximize return, while (2) executing diverse skills.
</p>
<div class="hero-body" style="display: flex; flex-direction: column; align-items: center; justify-content: center;">
<img src="./static/images/qdac.jpg" height="100%">
<h2 class="subtitle has-text-centered">
<span class="dnerf">QDAC</span> seamlessly unifies <span style="color: #e1144b;">performance</span> and <span style="color: #0053d6;">behavior</span> critics using <b>constrained optimization</b> to</br>(1) <span style="color: #e1144b;">maximize return</span>, while (2) <span style="color: #0053d6;">executing diverse skills</span>.
<span class="dnerf">QDAC</span> seamlessly unifies <span style="color: #e1144b;">performance</span> and <span style="color: #0053d6;">behavior</span> critics using <b>constrained optimization</b> to<br/>(1) <span style="color: #e1144b;">maximize return</span>, while (2) <span style="color: #0053d6;">executing diverse skills</span>.
</h2>
</div>
</div>
Expand Down

0 comments on commit 234d04b

Please sign in to comment.