Differences

This shows you the differences between two versions of the page.

--- news [2019/12/06 06:25]
IIS Webadmin
+++ news [2019/12/14 06:25]
IIS Webadmin
@@ Line 5: / Line 5: @@
 <HTML><table><tr valign="top"><td>2019-10-10</td><td>IIS guest <span style="font-weight:bold">Georg Martius</span>, Max Planck Institute for Intelligent Systems,
     Tübingen,  gives an invited colloquium <i>Control What You Can – Intrinsically Motivated
-    Hierarchical RL</i> at IIS Colloquium. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem11', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem11Abstract" style="display:none">I will present present a hierarchical reinforcement
+    Hierarchical RL</i> at IIS Colloquium. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem12', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem12Abstract" style="display:none">I will present present a hierarchical reinforcement
     learning agent that is intrinsically motivated to learn how to
     control its observation space in the fastest possible manner by
@@ Line 18: / Line 18: @@
     scratch.</blockquote></td></tr><tr valign="top"><td>2019-10-10</td><td>IIS guest <span style="font-weight:bold">Georg Martius</span>, Max Planck Institute for Intelligent Systems,
     Tübingen,  gives an invited colloquium <i>Self-Organization of Behavior in Autonomous Robot
-    Development</i> at <a href="https://www.uibk.ac.at/informatik/forschung/lunchtime-seminar/index.html.en">IFI Lunchtime Seminar</a>. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem12', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem12Abstract" style="display:none">I am studying the question how robots can autonomously
+    Development</i> at <a href="https://www.uibk.ac.at/informatik/forschung/lunchtime-seminar/index.html.en">IFI Lunchtime Seminar</a>. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem13', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem13Abstract" style="display:none">I am studying the question how robots can autonomously
     develop skills.  Considering children, it seems natural that they
     have their own agenda. They explore their environment in a playful
@@ Line 59: / Line 59: @@
     Playing and Autonomous Testing</i> at <a href="https://www.standort-tirol.at/page.cfm?vpath=veranstaltungen&amp;genericpageid=23118">Taste Digitalization – CAMPUS:digi:TOUR (4)</a>, ICT, Universität Innsbruck.</td></tr><tr valign="top"><td>2018-10-05</td><td>Justus Piater gives an invited talk <i>Perception of Abstract Concepts</i> at <a href="http://iros2018-uvsp.org/">Unconventional Sensing and
     Processing for Robotic Visual Perception</a>, Madrid, Spain.</td></tr><tr valign="top"><td>2018-09-21</td><td>Justus Piater gives an invited talk <i>Concept Learning For Robot Intelligence</i> at <a href="https://www.uibk.ac.at/congress/quantum-machine-learning-plus/index.html.en">Conference
-    on Quantum Machine Learning Plus</a>, Innsbruck, Austria.</td></tr><tr valign="top"><td>2018-07-17</td><td>Simon Haller and Nicolas Stolz appear in the media: <a href="https://www.uibk.ac.at/informatik/news/medienspiegel/nachlesen/2018_07_17_roboter_zu_besuch_in_liechtenstein-liechtensteiner-vaterland.pdf">
+    on Quantum Machine Learning Plus</a>, Innsbruck, Austria. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem40', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem40Abstract" style="display:none">Current robot learning, and machine learning in general,
+    requires carefully-engineered setups (environments, objective
+    functions, training data, etc.) for learning to succeed.
+    Perception and action spaces are specially crafted to meet the
+    requirements of the learning objective, which is specified in
+    advance.  How can we construct robot learning systems that can
+    learn in an open-ended fashion, acquire skills not foreseen by its
+    designers, and scale up to virtually unlimited levels of
+    complexity?  I argue that a key to achieving this lies in the
+    robot's ability to learn abstract concepts that can be reused as a
+    basis for future learning, both in autonomous exploration and for
+    teaching by humans.</blockquote></td></tr><tr valign="top"><td>2018-07-17</td><td>Simon Haller and Nicolas Stolz appear in the media: <a href="https://www.uibk.ac.at/informatik/news/medienspiegel/nachlesen/2018_07_17_roboter_zu_besuch_in_liechtenstein-liechtensteiner-vaterland.pdf">
     Roboter zu Besuch in Liechtenstein
     </a>.</td></tr><tr valign="top"><td>2018-07-03</td><td>Simon Haller teaches a tutorial <i>Wie kann ein Roboter lernen, die Welt zu verstehen?</i> at Campustag BORG Schwaz, Universität Innsbruck.</td></tr><tr valign="top"><td>2018-06-30</td><td>Philipp Zech, Barry Ridge, and Emre Ugur co-organize the <a href="https://afford.gitlab.io/rss-workshop/">1st
@@ Line 65: / Line 76: @@
     Robotics</a>.</td></tr><tr valign="top"><td>2018-06-30</td><td>Justus Piater gives an invited talk <i>Objects, Parts, Functions, Affordances, Actions, and Tasks</i> at <a href="https://afford.gitlab.io/rss-workshop/">1st
     International Workshop on Computational Models of Affordance in
-    Robotics</a>, Pittsburgh, PA, USA.</td></tr><tr valign="top"><td>2018-06-12</td><td>Justus Piater, Simon Haller, and Olaf Köhler appear in the media: <a href="https://www.uibk.ac.at/newsroom/music-globe-und-cubeat.html.de">
+    Robotics</a>, Pittsburgh, PA, USA. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem45', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem45Abstract" style="display:none">Human-made objects are constructed with functional
+    parts.  These are closely related to affordances; e.g., a handle
+    is a part whose function is to be grasped and that affords
+    grasping.  It is these affordances that robots need to detect to
+    perform tasks flexibly using unknown objects.  Since affordances
+    are often associated with parts, it makes sense for the
+    segmentation of unknown objects into parts for affordance
+    detection to be driven by known affordances.  Then, the
+    suitability of an object for a given task is often characterized
+    by the combination of the actions its various parts afford. For
+    example, to drive a nail into the wall you would like an object
+    that possesses a part that affords pounding and another that
+    affords grasping.  I will discuss methods that allow robots to
+    learn to segment objects in terms of their affordances, and to
+    learn the suitability of objects for specific tasks as a function
+    of their combination of affordances.</blockquote></td></tr><tr valign="top"><td>2018-06-12</td><td>Justus Piater, Simon Haller, and Olaf Köhler appear in the media: <a href="https://www.uibk.ac.at/newsroom/music-globe-und-cubeat.html.de">
     Music Globe und Cubeat
     </a>.</td></tr><tr valign="top"><td>2018-06-12</td><td>Simon Haller teaches a tutorial <i>Robotics im Klassenzimmer</i> at <a href="https://www.uibk.ac.at/informatik/schule/inday-teachers/index.html.de">InDay Teachers 2018</a>, Universität Innsbruck.</td></tr><tr valign="top"><td>2018-04-24</td><td>Justus Piater gives an invited talk <i>Intelligente Roboter – Freund, Feind oder

User Tools

Site Tools

Differences

Page Tools