Differences

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

--- start [2025/08/29 17:00]
Justus Piater
+++ start [2026/11/02 14:12] (current)
Simon Haller-Seeber [Working With Us]
@@ Line 7: / Line 7: @@
 We seek to answer the question: //How can we enable robots to acquire the knowledge and understanding they require to interact sensibly with unstructured environments? //
-Our research addresses complete perception-action loops, from computer vision to grasping and manipulation, using reactive algorithms and/or cognitive models.  Much of our work uses machine learning to enable robots to synthesize and improve complex and robust sensorimotor behavior with experience. Related areas of interest include human-robot interaction, image and video analysis, and visual neuroscience.
+Our research addresses complete perception-action loops, from computer vision to grasping and manipulation, using reactive algorithms and/or cognitive models.  Much of our work uses machine learning to enable robots to synthesize and improve complex and robust sensorimotor behavior with experience.
+===== Working With Us =====
-===== Working With Us =====
   * Check our thesis topics for [[theses/start|Bachelor and Master students]].
   * [[:jobs:#notice_for_non-eueea_prospective_master_students|Notice]] for non-[[https://en.wikipedia.org/wiki/EU|EU]]/[[https://en.wikipedia.org/wiki/European_Economic_Area|EEA]] prospective Master students
-  * We currently do not have any open PhD positions.
+  * We do not currently have any open PhD positions. If you'd like to work with us starting in summer 2027, apply via the [[https://ellis.eu/news/ellis-phd-program-call-for-applications-2026|ELLIS PhD Program]] by 2026-10-31.
-<html><!--We currently have [[jobs|one open PhD position]].--></html>
-<html><!--
-//Bachelor students://{{ ::mobile-manipulator-crop.jpg?nolink&100|}} **Want to create artificial intelligence for autonomous robots?** Want to join our [[@/uibk/piater/courses/RC_Poster2.pdf |interdisciplinary LFUI team]] to compete in the 2020 [[https://www.robocup.org/leagues/16|RoboCup@Work]] competition? Take the [[https://lfuonline.uibk.ac.at/public/lfuonline_lv.details?sem_id_in=19W&lvnr_id_in=703135|Introduction to Robotics]] course in the winter semester 2019-20!
---></html>
 <html>
@@ Line 25: / Line 20: @@
 </html>
-{{ ::iis_retreat_2024.jpg?direct&900 |}}\\
+{{ ::iis-group-2026.jpg?direct&1000 |}}\\
 <html>
 <div style="text-align: center;">
-  <p>Group picture taken at our 2024 retreat at Meissner Haus.</p>
+  <p>Group picture taken at our 2026 retreat at Maria Waldrast.</p>
 </div>
 </html>
@@ Line 34: / Line 29: @@
 ===== News =====
-<HTML><table><tr valign="top"><td>2025-07-15</td><td>Justus Piater gives an invited keynote <i>Making robots learn to perceive and act with
+<HTML><table><tr valign="top"><td>2026-09-27</td><td>Justus Piater co-organizes the <a href="https://worldmodelworkshop.github.io/">RoBoWoMo:
-    understanding</i> at <a href="https://www.uibk.ac.at/en/congress/multibody2025/">The
+    Bridging the Gap between Neural and Symbolic World Models for
-th ECCOMAS Thematic Conference on Multibody Dynamics</a>, Innsbruck. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem0', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem0Abstract" style="display:none">The flexibility and robustness of current robots is
+    Robot Planning, Reasoning, and Action</a>.</td></tr><tr valign="top"><td>2026-08-23 – 2026-08-28</td><td>David J. Crandall, Dirk Nowotka, Justus Piater, and Katrin Schöning-Stierand co-organize the <a href="https://www.dagstuhl.de/en/seminars/seminar-calendar/seminar-details/26351">Dagstuhl
-    limited by their lack of understanding of their environment. For
+    Seminar 26351 on Digital Transformation of Research and Digital
-    this reason, most robots operate in controlled environments.
+    Centers in the Science Landscape</a>.</td></tr><tr valign="top"><td>2026-05-20</td><td>Simon Haller-Seeber appears in the media: <a href="https://www.digital.tirol/page.cfm?vpath=news&amp;rnpageid=42902">
-    Machine learning can circumvent modeling problems but introduces
+        Digitale Bildung aktiv erleben: Der RoboCupJunior Austrian Open 2026 in Tirol</a>. (Interview im Rahmen des RoboCup Junior Austrian Open)</td></tr><tr valign="top"><td>2026-05-07</td><td>Justus Piater is a panelist at a public discussion on <a href="https://www.uibk.ac.at/de/medien-kommunikation/kommunikation/intern/">Invited
-    new problems of generalizing from examples. How can robots acquire
+    expert interview on “Human-robotics relations: What does the
-    understanding (of structure, function, causality, etc.) that
+    future hold?”</a>, Innsbruck. (Public event organized by the Department of Media, Society and Communication, Universität Innsbruck)</td></tr><tr valign="top"><td>2026-04-15</td><td>Justus Piater gives an invited talk <i>Some Latest Results in Robot Learning of Structure by
-    allows them to generalize from sparse experience? Motivated by
+    Interaction</i> at <a href="https://airov.at/2026/workshop/TactileRobotics.html">Austrian
-    shortcomings of current machine-learning methods, I will argue
+    Robotics Workshop</a>, Leoben. (Annual workshop of the GMAR; at AIRoV 2026)</td></tr><tr valign="top"><td>2026-04-08</td><td>Simon Haller-Seeber and Justus Piater co-organize the <a href="https://robocupjunior.at/">RCJ2026 - Robocup Junior Austrian Open: 08.-10.4.2026</a>.</td></tr><tr valign="top"><td>2026-04-08</td><td>Justus Piater gives an invited talk <i>Generative KI: Funktionsweise, Möglichkeiten und
-    that &quot;understanding&quot; is a meaningful notion in AI that reaches
+    Grenzen</i> at SchulleiterInnen-Tagung „IT-Sicherheit und KI-Einsatz in
-    beyond prediction and control. I will discuss examples of our
+    der Schule“, Pädagogische Hochschule Tirol. (Veranstaltung für Schulleitungen der Bildungsdirektion für Tirol) <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem7', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem7Abstract" style="display:none">Generative KI-Systeme wie ChatGPT und Midjourney haben
-    recent work on learning visual relational concepts, extrapolation
+    die Welt im Sturm erobert. Wie können wir sie produktiv nutzen?
-    of learned movements beyond the training distribution, learning of
+    Wie können wir mit den Herausforderungen umgehen, die durch sie
-    symbolic concepts and rules, and structure-driven skill learning
+    entstehen? Ich werde diese Fragen von der technischen Seite her
-    from sensorimotor experience. Our long-term objective is to
+    angehen und einen Eindruck davon vermitteln, wie diese Systeme
-    improve abstraction, generalization, robustness, and ultimately
+    funktionieren. Daraus ergeben sich ein realistisches Verständnis
-    explainability of robot perception and action.</blockquote></td></tr><tr valign="top"><td>2025-07-11</td><td>Justus Piater and Alejandro Agostini give an invited talk <i>Learning Symbols and Abstractions in Robot Planning</i> at <a href="https://www.unibz.it/en/events/abstraction-language-science-engineering">Abstraction:
+    für ihre Möglichkeiten und Grenzen sowie einige grundlegende
-    Language - Science - Engineering</a>, Bolzano. (International Workshop)</td></tr><tr valign="top"><td>2025-04-25</td><td>Simon Haller-Seeber gives an invited talk <i>AI-powered tools for real-time transcription and translation in action: A self-hosted open-source framework for digital spaces.</i> at <a href="https://www.uibk.ac.at/de/medien/veranstaltungen/tagungen/medien-wissen-bildung-2025/">Medien - Wissen - Bildung 2025: Streifzüge an den Nahtstellen von Medien, Bildung und Philosophie</a>, Universität Innsbruck.</td></tr><tr valign="top"><td>2025-04-10</td><td>Simon Haller-Seeber gives an invited keynote <i>Hands-on Approaches to Software, Robotics, and AI: Exploring Experiments and Science in Action</i> at <a href="https://iis.uibk.ac.at/public/simon/ECER-2025/">European Conference on Educational Robotics (ECER 2025)</a>, HTL Anichstrasse.</td></tr><tr valign="top"><td>2025-04-10</td><td>Simon Haller-Seeber and Christopher Kelter teach a tutorial <i>Was sind Roboter, was macht eine KI? Entwickle deine eigene KI und programmiere unsere Minibots</i> at Campustag BG/BRG Sillgasse, Universität Innsbruck.</td></tr><tr valign="top"><td>2025-04-03</td><td>Justus Piater gives an invited keynote <i>What is AI really?</i> at <a href="https://www.uibk.ac.at/events/2025/04/03/ai-in-law-and-practice-regional-perspectives-on-european-rules">Cross-Border
+    Empfehlungen für den Umgang mit ihnen im Schulbetrieb.</blockquote></td></tr><tr valign="top"><td>2026-03-31</td><td>Samuele Tosatto gives an invited keynote <i>Accelerating Reinforcement Learning with Off-Policy Data: Promises, Pitfalls, and Future Directions</i> at <a href="https://rl4aa.github.io/RL4AA26/">Reinforcement Learning For Autonomous Accelerators 2026</a>, Liverpool. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem8', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem8Abstract" style="display:none">Reinforcement learning is a promising technique for solving complex control problems in real-world physical systems, such as robotics, plasma stabilization, and particle accelerators. However, RL is often data-hungry, and its classic on-policy formulation is often inefficient, as it disallows data reuse, and unsafe, as it requires the agent to interact with the environment from scratch.
-    Seminar: AI in Law and Practice: Regional Perspectives on European
+Off-policy reinforcement learning offers a more appealing paradigm by enabling the reuse of historical data and the utilization of safe, external behavior sources (such as human operator logs). However, this flexibility comes at a cost: off-policy learning introduces significant theoretical instabilities. In this talk, we will analyze some fundamental difficulties in off-policy reinforcement learning, both in value and policy learning, explore the algorithmic landscape that tames them, and see the future direction in which the field is moving. </blockquote></td></tr><tr valign="top"><td>2025-11-19</td><td>Justus Piater gives an invited talk <i>Structural Understanding – The Grand Challenge of Robot
-    Rules</a>, Universität Innsbruck.</td></tr><tr valign="top"><td>2025-03-06</td><td>Justus Piater teaches a tutorial <i>Funktionsweise, Möglichkeiten und Grenzen von KI</i> at Lieber gleich berechtigt als später: Hallo KI, hilfst du
+    Learning</i> at <a href="https://elliit.se/news-and-events/focus-period-lund-2025/symposium/">ELLIIT Focus Period Symposium: Robot Learning</a>, Lund University. (The ELLIIT Focus Period Symposium is the highlight of the five-week focus period, during which young international scholars, ELLIIT researchers and other well-established international academics gather in Lund to work together on joint research challenges.) <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem9', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem9Abstract" style="display:none">AI has made great progress in recent years, and the
-    uns bei der Gleichstellung?!, Tiroler Bildungsinstitut Grillhof Vill. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem6', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem6Abstract" style="display:none">Künstliche Intelligenz&quot; bezeichnet derzeit Systeme, die
+    sophistication of robots has been rising with costs falling. Yet,
-    auf maschinellem Lernen (ML) basieren. Dieser Workshop führt in
+    the capabilities of AI-enabled robots are not keeping pace. I
-    die Grundlagen des ML ein, mit besonderem Augenmerk auf neuronale
+    argue that this is due to a lack of structural understanding by
-    Netze, der derzeit populärsten ML-Technologie. Darauf aufbauend
+    current AI systems. I will discuss several lines of research in my
-    vermittelt er ein Grundverständnis für prinzipielle Möglichkeiten
+    lab that seek to enable robots to generalize better and learn
-    und Grenzen von ML und warum es mit derzeitigen Mitteln extrem
+    faster thanks to explicit notions of structure.</blockquote></td></tr><tr valign="top"><td>2025-09-18</td><td>Samuele Tosatto gives an invited keynote <i>Where are all the intelligent robots? A quest for efficiency in reinforcement learning</i> at <a href="https://sarl-plus.github.io/RL-Bootcamp2025/">Reinforcement Learning Bootcamp 2025</a>, Salzburg. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem10', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem10Abstract" style="display:none">As artificial intelligence reshapes our digital world at a breathtaking pace,
-    schwierig ist, Diskriminierung in KI-Systemen zu
+    a curious question arises: Where are all the real-world, intelligent robots?
-    verhindern.</blockquote></td></tr><tr valign="top"><td>2025-02-26</td><td>Simon Haller-Seeber and Christopher Kelter teach a tutorial <i>Programmieren eines autonomen Roboters anhand einer selbstentwickelten KI</i> at STAIR-Lab INNALP Workshop für das Gymnasium Ursulinenen Innsbruck, Universität Innsbruck.</td></tr><tr valign="top"><td>2025-02-26</td><td>Marko Zarić teaches a tutorial <i>Einführung in das Programmieren mit Microcontrollern</i> at STAIR-Lab INNALP Workshop für das Gymnasium Ursulinenen Innsbruck, Universität Innsbruck.</td></tr><tr valign="top"><td>2025-01-31</td><td>Justus Piater gives an invited lecture <i>High-Level AI For Autonomous Robots</i> at <a href="https://elias-ai.eu/event/ellis-vismac/">ELIAS-ELLIS-VISMAC
+    While we have mastered the generation of text, images, and video by leveraging vast web-scale datasets,
-    Winter School 2025</a>, Brunico, Italy. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem9', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem9Abstract" style="display:none">What barriers will have to be overcome to raise the
+    robotics still faces a fundamental data bottleneck. Reinforcement learning (RL) offers a compelling solution,
-    flexibility and robustness of autonomous robots to the next level?
+    enabling agents to learn autonomously by collecting their own experience.
-    I argue that current AI methods are unlikely to get there, no
+    However, the path to autonomy is often blocked by the staggering inefficiency of current RL algorithms,
-    matter how far we scale them up. Robots need to be able to
+    which can require millions of trials to master simple tasks. This talk embarks on a quest to tackle this
-    &quot;understand&quot; their environment in ways that reach beyond
+    efficiency problem head-on. I will argue that a crucial step toward unlocking the potential of
-    prediction and control. I will discuss examples of work (ours and
+    robot learning lies in a two-pronged approach: first, by developing statistically efficient
-    others') on learning visual relational concepts, extrapolation of
+    algorithms that can reuse data by leveraging sound off-policy techniques, and second,
-    learned movements beyond the training distribution, learning of
+    by designing better action representations for physical, real-world agents.
-    symbolic concepts and rules, and structure-driven skill learning
+    By making our algorithms more efficient and refining their core hypotheses,
-    from sensorimotor experience.</blockquote></td></tr></table></HTML>
+    we can accelerate the journey toward real embodied artificial intelligence.</blockquote></td></tr></table></HTML>
 [[news|Older News]]

User Tools

Site Tools

Differences

Page Tools