<div dir="ltr">







<p class=""><font face="arial, helvetica, sans-serif">This is just a friendly reminder that the paper submission deadline for this workshop is on April 10, 2014.<br></font><span style="font-family:arial,helvetica,sans-serif">We also have an exciting list of confirmed invited speakers. Make sure that you participate!<br>
</span><span style="font-family:arial,helvetica,sans-serif">-am</span></p>
<p class=""><br></p>
<p class=""><b><font face="arial, helvetica, sans-serif">The AAAI-14 Workshop on Sequential Decision-Making with Big Data<br></font></b><b><font face="arial, helvetica, sans-serif">held at the AAAI Conference on Artificial Intelligence (AAAI-14), Quebec City, Canada, July 27-28, 2014.</font></b></p>

<p class=""><span style="font-family:arial,helvetica,sans-serif">Workshop URL: </span><a href="https://sites.google.com/site/decisionmakingbigdata" style="font-family:arial,helvetica,sans-serif"><span class="">https://sites.google.com/site/decisionmakingbigdata</span></a><br>
</p>
<p class=""><span style="font-family:arial,helvetica,sans-serif">In the 21st century, we live in a world where data is abundant. We would like to use this data to make better decisions in many areas of life, such as industry, health care, business, and government. This opportunity has encouraged many machine learning and data mining researchers to develop tools to benefit from big data. However, the methods developed so far have focused almost exclusively on the task of prediction. As a result, the question of how big data can leverage decision-making has remained largely untouched.</span><br>
</p>
<p class=""><font face="arial, helvetica, sans-serif">This workshop is about decision-making in the era of big data. The main topic will be the complex decision-making problems, in particular the sequential ones, that arise in this context. Examples of these problems are high-dimensional large-scale reinforcement learning and their simplified version such as various types of bandit problems. These problems can be classified into three potentially overlapping categories:</font></p>

<p class=""><span style="font-family:arial,helvetica,sans-serif">1) Very large number of data-points. Examples: data coming from user clicks on the web and financial data. In this scenario, the most important issue is computational cost. Any algorithm that is super-linear will not be practical.</span></p>
<p class=""><span style="font-family:arial,helvetica,sans-serif">2) Very high-dimensional input space. Examples are found in robotic and computer vision problems. The only possible way to solve these problems is to benefit from their regularities.</span><span style="font-family:arial,helvetica,sans-serif"><br>
</span></p><p class=""><span style="font-family:arial,helvetica,sans-serif">3) Partially observable systems. Here the immediate observed variables do not have enough information for accurate decision-making, but one might extract sufficient information by considering the history of observations. If the time series is projected onto a high-dimensional representation, one ends up with problems similar to 2. </span><font face="arial, helvetica, sans-serif"><br>
</font></p>
<p class=""><br></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Topics<br></font></b><span style="font-family:arial,helvetica,sans-serif">Some potential topics of interest are:<br></span><span style="font-family:arial,helvetica,sans-serif">- Reinforcement learning algorithms that deal with one of the aforementioned categories<br>
</span><span style="font-family:arial,helvetica,sans-serif">- Bandit problems with high-dimensional action space<br></span><span style="font-family:arial,helvetica,sans-serif">- Challenging real-world applications of sequential decision-making problems that can benefit from big data. Example domains include robotics, adaptive treatment strategies for personalized health care, finance, recommendation systems, and advertising.</span></p>

<p class=""><font face="arial, helvetica, sans-serif"><br></font></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Format<br></font></b><span style="font-family:arial,helvetica,sans-serif">The workshop will be a one-day meeting consisting of invited talks, oral and poster presentations from participants, and a final panel-driven discussion.</span></p>

<p class=""><font face="arial, helvetica, sans-serif"><br></font></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Attendance<br></font></b><span style="font-family:arial,helvetica,sans-serif">We expect about 30-50 participants from invited speakers, contributed authors, and interested researchers.</span></p>

<p class=""><font face="arial, helvetica, sans-serif"><br></font></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Submission<br></font></b><span style="font-family:arial,helvetica,sans-serif">We invite researchers from different fields of machine learning (e.g., reinforcement learning, online learning, active learning), optimization, systems (distributed and parallel computing), as well as application-domain experts (from e.g., robotics, recommendation systems, personalized medicine, etc.) to submit an extended abstract (maximum 4 pages in AAAI format) of their recent work to <a href="mailto:decision.making.big.data@gmail.com">decision.making.big.data@gmail.com</a>. Accepted papers will be presented as posters or contributed oral presentations.</span></p>

<p class=""><br></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Important Dates<br></font></b><span style="font-family:arial,helvetica,sans-serif">Paper Submission: </span><b style="font-family:arial,helvetica,sans-serif">April 10, 2014<br>
</b><span style="font-family:arial,helvetica,sans-serif">Notification of Acceptance: May 1, 2014<br></span><span style="font-family:arial,helvetica,sans-serif">Camera-Ready Papers: May 15, 2014<br></span><span style="font-family:arial,helvetica,sans-serif">Date of Workshop: July 27 or 28, 2014</span></p>

<p class=""><br></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Invited Speakers<br></font></b><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://www.cs.cmu.edu/~ebrun/" style="font-family:arial,helvetica,sans-serif"><span class="">Emma Brunskill</span></a><span style="font-family:arial,helvetica,sans-serif"> (Carnegie Mellon University)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://research.microsoft.com/en-us/people/lihongli/" style="font-family:arial,helvetica,sans-serif"><span class="">Lihong Li</span></a><span style="font-family:arial,helvetica,sans-serif"> (Microsoft Research)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://webdocs.cs.ualberta.ca/~sutton/" style="font-family:arial,helvetica,sans-serif"><span class="">Richard Sutton</span></a><span style="font-family:arial,helvetica,sans-serif"> (University of Alberta)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- And a few others to be confirmed later.</span></p>
<p class=""><br></p>
<p class=""><b><font face="arial, helvetica, sans-serif">Organizing Committee<br></font></b><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://academic.sologen.net/Home.html" style="font-family:arial,helvetica,sans-serif"><span class="">Amir-massoud Farahmand</span></a><span style="font-family:arial,helvetica,sans-serif"> (Carnegie Mellon University)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://www.cs.mcgill.ca/~amsb/" style="font-family:arial,helvetica,sans-serif"><span class="">André M.S. Barreto</span></a><span style="font-family:arial,helvetica,sans-serif"> (Brazilian National Laboratory for Scientific Computing (LNCC))<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://chercheurs.lille.inria.fr/~ghavamza/my_website/About_Me.html" style="font-family:arial,helvetica,sans-serif"><span class="">Mohammad Ghavamzadeh</span></a><span style="font-family:arial,helvetica,sans-serif"> (Adobe Research and INRIA Lille - Team SequeL)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://www.cs.mcgill.ca/~jpineau/" style="font-family:arial,helvetica,sans-serif"><span class="">Joelle Pineau</span></a><span style="font-family:arial,helvetica,sans-serif"> (McGill University)<br>
</span><span style="font-family:arial,helvetica,sans-serif">- </span><a href="http://www.cs.mcgill.ca/~dprecup/" style="font-family:arial,helvetica,sans-serif"><span class="">Doina Precup</span></a><span style="font-family:arial,helvetica,sans-serif"> (McGill University)</span></p>
</div>