Skip to content

Youngdolee/homepage #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
leeyngdo merged 3 commits into from
Feb 24, 2025
Merged

Youngdolee/homepage #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e3c5031
Update Project Page
leeyngdo Feb 24, 2025
fc07428
Delete DS_Store
leeyngdo Feb 24, 2025
5441699
Delete DS_Store
leeyngdo Feb 24, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 20 additions & 0 deletions docs/dataset/css/bulma.min.css
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,21 @@
.bibtex-container{
flex-grow:1;
margin:0 auto
;position:relative;
width:auto
}

.bibtex pre{
-webkit-overflow-scrolling: touch;
overflow-x:auto;
padding:1.25em 1.5em;
white-space:pre;
word-wrap:normal;
font-family: "Courier", monospace;
background-color: #f4f4f4;
text-align: left;
}

.hero{
align-items:stretch;
display:flex;
Expand All @@ -10,4 +28,6 @@
}





12 changes: 11 additions & 1 deletion docs/dataset/css/style.css
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,8 +6,18 @@ h2, h3, h4, h5, a, p, span, body {font-weight: normal; font-family: "Google Sans
.header-menu {background-color: #efeff3; width: 100%; padding: 16px 0;}
.header-menu-content {max-width: 960px; margin: auto;}
.header-menu-item {display: inline-block; margin-left: 16px; margin-right: 16px; font-size: 1.2em;}
.links {width: 100%; margin: auto; text-align: center; padding-top: 8px;}

.links {width: 120%; margin: auto; text-align: center; padding-top: 8px; margin-left: -10%;}
.links a {margin-left: 8px;}
.links br {display: none;}

@media (max-width: 900px) {
.links {width: 100%; margin: auto; text-align: center; padding-top: 8px;}
.links br {display: block;}
.links a {margin-top: 8px; }
}


.content {max-width: 960px; margin: auto; margin-top: 48px; margin-bottom: 64px;}
a, h2 {color: rgb(100, 142, 246); text-decoration: none;}
a:hover {color: #fa6d6d;}
Expand Down
56 changes: 29 additions & 27 deletions docs/dataset/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@
<a href="../">Home</a>
</div>
<div class="header-menu-item noselect">
<a href="https://github.com/nicklashansen/simbav2">Code</a>
<a href="https://github.com/dojeon-ai/SimbaV2">Code</a>
</div>
<div class="header-menu-item noselect">
<a href="." class="disabled">Dataset</a>
Expand All @@ -51,19 +51,26 @@
<h2>Dataset</h2>
<div class="figure-caption">
<p>
We open-source all data corresponding to the 80-task and 30-task datasets used in our multi-task experiments. They can be downloaded below. The two datasets contain 545M and 345M transitions, respectively.
We're open-sourcing $57$-task datasets from the <span class="bold simbav2">SimbaV2</span> agent's replay buffers, available for download below. We hope this release encourages other research groups to share their datasets and checkpoints, driving collaboration and progress.
</p>
</div>
<div style="width: 100%; margin: auto; text-align: center; padding-top: 8px; font-size: 1.2em;">
<span class="bold">Click to Download:</span><br class="break"/>
</div><br>
<div class="links">
<span class="bold">Download:</span><a href="https://huggingface.co/datasets/nicklashansen/simbav2" class="btn btn-blue">All</a><a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt80" class="btn">80 tasks</a><a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt30" class="btn">30 tasks</a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2" class="btn btn-blue">🤗 Hugging Face</a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/mujoco.pickle" class="btn ">MuJoCo (5 tasks)</a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/dmc.pickle" class="btn ">DMControl (28 tasks)</a><br class="break"/>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/myosuite.pickle" class="btn ">MyoSuite (10 tasks)</a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/hbench.pickle" class="btn ">HumanoidBench (14 tasks)</a>
</div>
</div>
<div class="hr"></div>
<div>
<h2>Overview</h2>
<div class="figure-caption">
<p>
We release two multi-task datasets with data from <span class="bold">80</span> and <span class="bold">30</span> tasks, respectively. The datasets are summarized below.
We release a total of $57$ single-task expert datasets with transition data from the replay buffers of single-task <span class="bold simbav2">SimbaV2</span> agents. The domains consist of <span class="bold">MuJoCo (5)</span>, <span class="bold">DMControl (28)</span>, <span class="bold">MyoSuite (10)</span>, and <span class="bold">HumanoidBench (14)</span>, encompassing a variety of locomotion and manipulation tasks with varying levels of complexity. The dataset details are summarized below.
</p>
</div>
<table class="models" style="min-width: 1em;">
Expand All @@ -89,19 +96,19 @@ <h2>Overview</h2>
5
</td>
<td style="font-size: 0.8em;">
4
5
</td>
<td style="font-size: 0.8em;">
690k
</td>
<td style="font-size: 0.8em;">
345M
171M
</td>
<td style="font-size: 0.8em;">
20GB
11.4GB
</td>
<td>
<a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt30"><i class="fa fa-download" style="font-size: 1em;"></i></a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/mujoco.pickle"><i class="fa fa-download" style="font-size: 1em;"></i></a>
</td>
</tr>
<tr class="models">
Expand All @@ -116,19 +123,19 @@ <h2>Overview</h2>
28
</td>
<td style="font-size: 0.8em;">
11
13
</td>
<td style="font-size: 0.8em;">
690k
</td>
<td style="font-size: 0.8em;">
345M
171M
</td>
<td style="font-size: 0.8em;">
20GB
</td>
<td>
<a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt30"><i class="fa fa-download" style="font-size: 1em;"></i></a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/dmc.pickle"><i class="fa fa-download" style="font-size: 1em;"></i></a>
</td>
</tr>
<tr class="models">
Expand All @@ -149,13 +156,13 @@ <h2>Overview</h2>
690k
</td>
<td style="font-size: 0.8em;">
345M
171M
</td>
<td style="font-size: 0.8em;">
20GB
14.3GB
</td>
<td>
<a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt30"><i class="fa fa-download" style="font-size: 1em;"></i></a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/myosuite.pickle"><i class="fa fa-download" style="font-size: 1em;"></i></a>
</td>
</tr>
<tr class="models">
Expand All @@ -176,29 +183,24 @@ <h2>Overview</h2>
690k
</td>
<td style="font-size: 0.8em;">
345M
171M
</td>
<td style="font-size: 0.8em;">
20GB
11.1GB
</td>
<td>
<a href="https://huggingface.co/datasets/nicklashansen/simbav2/tree/main/mt30"><i class="fa fa-download" style="font-size: 1em;"></i></a>
<a href="https://huggingface.co/datasets/joonleesky/simbaV2/blob/main/hbench.pickle"><i class="fa fa-download" style="font-size: 1em;"></i></a>
</td>
</tr>
</tbody>
</table>
<div class="figure-caption">
<p>
These datasets are obtained from the replay buffers of <span class="bold">240</span> single-task <span class="bold simbav2">SimbaV2</span> agents, and thus contain a wide variety of behaviors ranging from random to expert policies. Multi-task agents trained on the above datasets, as well as checkpoints for each of the single-task agents, are available <a href="https://www.simbav2.com/models">here</a>.
</p>
</div>
</div>
<div class="hr"></div>
<div>
<h2>Tasks</h2>
<div class="figure-caption">
<p>
We list all 57 tasks in the dataset below, along with <span class="bold simbav2">SimbaV2</span> policy visualizations and summary statistics.
We list all $57$ tasks in the dataset below, along with <span class="bold simbav2">SimbaV2</span> policy visualizations and: summary statistics.
</p>
</div>
<!-- Mujoco Videos -->
Expand Down Expand Up @@ -248,7 +250,7 @@ <h2>Tasks</h2>
<!-- MuJoCo Table -->
<div class="wrap-collabsible">
<input id="collapsible-mujoco" class="toggle" type="checkbox">
<label for="collapsible-mujoco" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">MuJoCo Full List ($5$ tasks)</label>
<label for="collapsible-mujoco" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">MuJoCo: Summary Statistics ($5$ tasks)</label>
<div class="collapsible-mujoco-content">
<div class="content-inner">
<table class="models">
Expand Down Expand Up @@ -552,7 +554,7 @@ <h2>Tasks</h2>
<!-- DMC Table -->
<div class="wrap-collabsible">
<input id="collapsible-dmc" class="toggle" type="checkbox">
<label for="collapsible-dmc" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">Deepmind Control Suite Full List ($28$ tasks)</label>
<label for="collapsible-dmc" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">DMControl: Summary Statistics ($28$ tasks)</label>
<div class="collapsible-dmc-content">
<div class="content-inner">
<table class="models">
Expand Down Expand Up @@ -1098,7 +1100,7 @@ <h2>Tasks</h2>
<!-- MyoSuite Table -->
<div class="wrap-collabsible">
<input id="collapsible-myosuite" class="toggle" type="checkbox">
<label for="collapsible-myosuite" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">MyoSuite Full List ($10$ tasks)</label>
<label for="collapsible-myosuite" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">MyoSuite: Summary Statistics ($10$ tasks)</label>
<div class="collapsible-myosuite-content">
<div class="content-inner">
<table class="models">
Expand Down Expand Up @@ -1384,7 +1386,7 @@ <h2>Tasks</h2>
<!-- HBench Table -->
<div class="wrap-collabsible">
<input id="collapsible-hbench" class="toggle" type="checkbox">
<label for="collapsible-hbench" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">HumanoidBench Full List ($14$ tasks)</label>
<label for="collapsible-hbench" class="lbl-toggle" style="font-size: min(1.2em, 4.5vw);">HumanoidBench: Summary Statistics ($14$ tasks)</label>
<div class="collapsible-hbench-content">
<div class="content-inner">
<table class="models">
Expand Down
1 change: 1 addition & 0 deletions docs/dataset/js/bulma-carousel.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1036,6 +1036,7 @@ var defaultOptions = {
autoplaySpeed: 3000
};


var Autoplay = function (_EventEmitter) {
_inherits(Autoplay, _EventEmitter);

Expand Down
Binary file added docs/dataset/videos/.DS_Store
View file
Open in desktop
Binary file not shown.
39 changes: 25 additions & 14 deletions docs/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
<meta property="og:title" content="SimbaV2:">
<meta property="og:description" content="Hyperspherical Normalization for Scalable Deep Reinforcement Learning">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" href="./dataset/css/bulma.min.css">
<link rel="stylesheet" href="style.css">
<link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css">
Expand Down Expand Up @@ -31,7 +32,7 @@ <h3><a class="bold default-color">Under Review</span><br/></h3>
<tr>
<td>
<h4>
<a href="https://joonleesky.github.io" class="nobreak">Hojoon Lee</a><sup>1, 2</sup>$\dagger$,&ensp;
<a href="https://joonleesky.github.io" class="nobreak">Hojoon Lee</a><sup>1</sup>$\dagger$,&ensp;
<a href="https://leeyngdo.github.io/" class="nobreak">Youngdo Lee</a><sup>1</sup>$\dagger$,&ensp;
<a href="https://takuseno.github.io/" class="nobreak">Takuma Seno</a><sup>2</sup>&ensp;
<a href="https://i-am-proto.github.io" class="nobreak">Donghu Kim</a><sup>1</sup>&ensp;<br/>
Expand All @@ -52,7 +53,7 @@ <h4>
</tbody>
</table>
<div class="links" style="margin-top: -20px;">
<a href="https://arxiv.org/abs/2502.15280" class="btn"><i class="fa">&#xf1c1;</i>&ensp;Paper</a><a href="https://github.com/dojeon-ai/SimbaV2" class="btn"><i class="fa fa-github"></i>&ensp;Code</a><a href="./dataset" class="btn"><i class="fa fa-database"></i> Dataset</a>
<a href="https://arxiv.org/abs/2502.15280" class="btn"><i class="fa">&#xf1c1;</i>&ensp;Paper</a><a href="https://github.com/dojeon-ai/SimbaV2" class="btn"><i class="fa fa-github"></i>&ensp;Code</a><a href="./dataset" class="btn"><i class="fa fa-database"></i>&ensp;Dataset</a>
</div>
<div class="content">
<!-- <div class="figure" style="height: 360px; background-image: url(images/overview.png);"></div> -->
Expand All @@ -69,7 +70,7 @@ <h4>
<div class="tldr-container">
<h2>TL;DR</h2>
<p class="tldr-content">
Stop worrying about algorithms, just change the network architecture to <span class="bold simbav2">SimbaV2</span>
Stop worrying about algorithms, just <span class="bold">change the network architecture to <span class="bold simbav2">SimbaV2</span></span>
</p>
</div>
<div class="hr"></div>
Expand Down Expand Up @@ -151,7 +152,7 @@ <h2>Scaling Network Size & UTD Ratio</h2>
<h2>Empiricial Analysis: Training Stability</h2>
<div class="figure-caption" style="margin-bottom: -30px;">
<p>
We track $4$ metrics during training to understand the learning dynamics of <span class="bold simbav2">SimbaV2</span> and <span class="bold simba">Simba</span>:
We track average return and $4$ metrics during training to understand the learning dynamics of <span class="bold simbav2">SimbaV2</span> and <span class="bold simba">Simba</span>:
<ul>
<li><strong>(a)</strong> Average normalized return across tasks</li><br>
<li><strong>(b)</strong> Weighted sum of $\ell_2$-norms of all intermediate features in critics</li><br>
Expand Down Expand Up @@ -281,7 +282,7 @@ <h2>SimbaV2 with Online RL</h2>
<h2>Paper</h2>
<span class="vbold"><span class="vbold simbav2">SimbaV2</span>: Hyperspherical Normalization for Scalable Deep Reinforcement Learning</span><br/>
<span class="italic">Hojoon Lee&ast;, Youngdo Lee&ast;, Takuma Seno, Donghu Kim, Peter Stone, Jaegul Choo</span><br/><br/>
<a href="https://arxiv.org">arXiv preprint</a><br/><br/>
<a href="https://arxiv.org/abs/2502.15280">arXiv preprint</a><br/><br/>
<div class="page" style="background-image: url(thumbnails/0.png);"></div>
<div class="page" style="background-image: url(thumbnails/1.png);"></div>
<div class="page" style="background-image: url(thumbnails/2.png);"></div>
Expand All @@ -290,21 +291,31 @@ <h2>Paper</h2>
<div class="page" style="background-image: url(thumbnails/5.png);"></div>
<div class="page" style="background-image: url(thumbnails/6.png);"></div>
<div class="page" style="background-image: url(thumbnails/7.png);"></div>
<div class="page" style="background-image: url(thumbnails/8.png);"></div>
<div class="page" style="background-image: url(thumbnails/9.png);"></div>
<div style="margin: auto; margin-top: 32px;">
<a href="https://arxiv.org/abs/2310.16828">View on arXiv</a>
<a href="https://arxiv.org/abs/2502.15280">View on arXiv</a>
</div>
</div>
<div class="hr"></div>
<div style="padding-bottom: 64px; text-align: center;">
<h2>Citation</h2>
<div class="bibtex-container bibtex">
<h2 class="title">Citation</h2>
<p class="citation">
If you find our work useful, please consider citing the paper as follows:
If you find our work useful, please consider citing the paper as follows:
</p>
<div id="bibtex-text" class="bibtexsection" onClick="window.getSelection().selectAllChildren(document.getElementById('bibtex-text'));">
</div>
</div>
<pre><code>@article{lee2025simbav2,
title={Hyperspherical Normalization for Scalable Deep Reinforcement Learning},
author={Hojoon Lee and Youngdo Lee and Takuma Seno and Donghu Kim and Peter Stone and Jaegul Choo},
journal={arXiv preprint arXiv:2502.15280},
year={2025},
}</code></pre>
</div>
<!-- <div class="container is-max-desktop content" style="padding-bottom: 64px; text-align: center;">
<h2 class="title">Citation</h2>
<p class="citation">
If you find our work useful, please consider citing the paper as follows:
</p>
<div class="bibtexsection">
</div>
</div> -->
<footer>
<a href="index.html#top"><i class="fa fa-arrow-up"></i><br/>Return to top</a>
<div style="padding-top: 48px;">
Expand Down
3 changes: 1 addition & 2 deletions docs/style.css
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ a:hover {color: #fa6d6d;}
.hr {width: 100%; height: 1px; margin: 48px 0; background-color: #d6dbdf;}
p {line-height: 1.4em; text-align: justify;}
.abstract {max-width: 90%; margin: auto;}
.citation {max-width: 90%; margin: auto;}
.citation {max-width: 95%; margin: auto; margin-bottom: 1em;}
.math {font-family: "Computer Modern Sans", sans-serif; font-style: italic;}
sub, sup {line-height: 0;}
.figure {width: 100%; min-height: 120px; margin: 2em 0; background-repeat: no-repeat; background-position: center; background-size: contain;}
Expand Down Expand Up @@ -70,7 +70,6 @@ a.btn-blue {background-color: #648ef6; color: white;}
a.btn-blue:hover {background-color: #fa6d6d;}
.header-menu-item a.disabled {color: rgb(47, 47, 47) !important; pointer-events: none; text-decoration: underline;}
a.link-disabled {color: #ccc; pointer-events: none;}
.bibtexsection {padding: 4px 16px; font-family: "Courier", monospace; font-size: 15px; white-space: pre; background-color: #f4f4f4; text-align: left;}
.noselect {-webkit-touch-callout: none; -webkit-user-select: none; -khtml-user-select: none; -moz-user-select: none; -ms-user-select: none; user-select: none;}
.bold {font-weight: bold;}
.vbold {font-weight: bolder;}
Expand Down
Binary file modified docs/thumbnails/0.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/1.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/2.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/3.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/4.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/5.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/6.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/thumbnails/7.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file removed docs/thumbnails/8.png
View file
Open in desktop
Binary file not shown.
Binary file removed docs/thumbnails/9.png
View file
Open in desktop
Binary file not shown.
Loading