<div dir="ltr"><p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">Dear
Colleagues,</span><span style="font-size:10pt"><br>
<br>
<span style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">It is our pleasure to invite you to submit your
research papers to the Special Issue on “New Horizons in Multiagent
Learning” in the journal of Autonomous Agents and Multi-Agent Systems. </span></span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"> </span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><a href="https://static.springer.com/sgw/documents/1644414/application/pdf/AGNT+-+CFP+-+New+Horizons+in+Multiagent+Learning.pdf" style="color:rgb(5,99,193)">https://static.springer.com/sgw/documents/1644414/application/pdf/AGNT+-+CFP+-+New+Horizons+in+Multiagent+Learning.pdf</a></span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"> </span></p>

<p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span class="gmail-m-94874502213628718gmail-im"><span style="font-size:10pt">--------</span></span></p>

<p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt"><br>
<span class="gmail-il">Learning</span><span class="gmail-m-94874502213628718gmail-im"> is
a critical component for autonomous agents so that they can discover novel
behaviors, adapt to a non-stationary world, and handle unanticipated events. As
virtual and physical agents become more common, these </span><span class="gmail-il">learning</span><span class="gmail-m-94874502213628718gmail-im"> agents
need to handle interactions not only with an environment, but also with other
agents. </span><span class="gmail-il">Multiagent learning</span><span class="gmail-m-94874502213628718gmail-im"> thus becomes an indispensable component
of building intelligent systems, even though much of the current </span><span class="gmail-il">learning</span><span class="gmail-m-94874502213628718gmail-im"> research
focuses on single-agent settings.</span><br>
<br>
Recent breakthroughs <span class="gmail-il">in</span> deep <span class="gmail-il">learning</span> have energized both academic and industrial
research labs, allowing <span class="gmail-il">new </span>types and difficulties
of complex tasks to be successfully learned. Some recent work has also shown
the great promise of applying deep reinforcement <span class="gmail-il">learning</span> techniques <span class="gmail-il">in</span> multi-agent settings, incentivizing the further
exploration of this avenue and the development of <span class="gmail-il">new</span> deep <span class="gmail-il">learning</span> <span class="gmail-il">multiagent</span> architectures.
This line of research is particularly important because there are many <span class="gmail-il">new</span> problems that can be explored and tackled by
deep <span class="gmail-il">multiagent learning</span>, from both empirical and
theoretical perspectives.<br>
<br>
<span class="gmail-m-94874502213628718gmail-im">This special issue aims to gather
novel </span>research articles, overview articles, and position
papers to highlight the recent interest and advances <span class="gmail-il">in</span> multi-agent <span class="gmail-il">learning</span>. For the latter two types of papers, contacting the
guest editors prior to submission is highly encouraged. For research articles,
relevant topics include, but are not limited to, the following:</span></p>

<p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt">* <span class="gmail-il">Learning</span> <span class="gmail-il">in</span> Markov games,
imperfect information games<br>
* Swarm intelligence/<span class="gmail-il">learning</span><br>
* Evolutionary methods<br>
* Emergent behaviors<br>
* <span class="gmail-il">Learning</span> <span class="gmail-il">in</span> the
presence of other strategic agents<br>
* Opponent modeling<br>
* <span class="gmail-il">Learning</span> to communicate<br>
* Mechanism design for <span class="gmail-il">learning</span> agents<br>
* Inter-agent teaching and transfer<br>
* Speedup methods for multi-agent <span class="gmail-il">learning</span><br>
* <span class="gmail-il">Learning</span> for heterogeneous multi-agent
systems<br>
* <span class="gmail-il">Learning</span> <span class="gmail-il">in</span> multi-robot
systems<br>
* Combining <span class="gmail-il">multiagent</span> <span class="gmail-il">learning</span> with
other areas (e.g., search, <span class="gmail-il">learning</span> from
demonstrations, etc.)<br>
<br>
Issue Editors<br>
Matthew E. Taylor, Borealis AI (<a href="mailto:matthew.taylor@borealisai.com" target="_blank" style="color:rgb(5,99,193)">matthew.taylor@borealisai.com</a>)<br>
Karl Tuyls, Deepmind & KU Leuven (<a href="mailto:karltuyls@google.com" target="_blank" style="color:rgb(5,99,193)">karltuyls@google.com</a>)<br>
<br>
Submissions and Reviews Procedures<br>
Special Issues are handled <span class="gmail-il">in</span> the normal way
via the online Editorial Manager system found at <a href="https://agnt.edmgr.com/" target="_blank" style="color:rgb(5,99,193)">https://agnt.edmgr.com</a>. Please
choose the article type “S.I. : <span class="gmail-il">New</span> <span class="gmail-il">Horizons</span> <span class="gmail-il">in</span> <span class="gmail-il">Multiagent</span> <span class="gmail-il">Learning</span>.” Special Issue articles should fulfill all the
standard requirements of any JAAMAS article. Authors should note that the same
criteria apply to articles <span class="gmail-il">in</span> Special Issues as
to regular articles. All papers will undergo the same rigorous AGNT review
process. Please refer to the JAMAAS website for detailed instructions on paper
submission: <a href="http://www.springer.com/computer/ai/journal/10458" target="_blank" style="color:rgb(5,99,193)">http://www.springer.com/computer/ai/journal/10458</a></span></p>

<p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt"> </span></p>

<p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt">Submission
deadline: March 4, 2019</span></p><p class="MsoNormal" style="background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 0.0001pt;font-size:12pt;font-family:Calibri,sans-serif"><span style="font-size:10pt"><br></span></p><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div>---------------------------------------------<br>Matt Taylor</div><div>BorealisAI.com<br><a href="http://eecs.wsu.edu/~taylorm" target="_blank">http://eecs.wsu.edu/~taylorm</a><br></div></div></div></div></div>