start
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
start [2020/12/10 06:25] IIS Webadmin |
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? // | 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]]. | * Check our thesis topics for [[theses/start|Bachelor and Master students]]. | ||
| - | * [[https://ellis.eu/|ELLIS]] PhD Program: [[https://ellis.eu/de/news/ellis-phd-program-call-for-applications|Call for Applications]] | + | * [[: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 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><!-- | ||
| - | //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> | <html> | ||
| Line 22: | Line 20: | ||
| </html> | </html> | ||
| - | + | {{ ::iis-group-2026.jpg?direct&1000 |}}\\ | |
| - | {{::dsc_0052.jpg?640|Group Picture}}\\ | + | <html> |
| - | Group picture taken at our retreat in Obergurgl | + | <div style="text-align: center;"> |
| + | <p>Group picture taken at our 2026 retreat at Maria Waldrast.</p> | ||
| + | </div> | ||
| + | </html> | ||
| ===== News ===== | ===== News ===== | ||
| - | <HTML><table><tr valign="top"><td>2020-11-19</td><td>Justus Piater gives an invited talk <i>Machine Learning in Robotics</i> at <a href="https://baiome.org/">bAIome PI Talk, Center for | + | <HTML><table><tr valign="top"><td>2026-09-27</td><td>Justus Piater co-organizes the <a href="https://worldmodelworkshop.github.io/">RoBoWoMo: |
| - | Biomedical AI, University Medical Center Hamburg-Eppendorf</a>, online. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem1', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem1Abstract" style="display:none">Machine Learning increasingly equips robots with | + | Bridging the Gap between Neural and Symbolic World Models for |
| - | learning capabilities and flexibility. This will enable them to | + | 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 |
| - | act purposefully in unstructured environments and to react to | + | Seminar 26351 on Digital Transformation of Research and Digital |
| - | unforeseen events. People can teach them intuitively to perform | + | 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&rnpageid=42902"> |
| - | tasks instead of having to program them in painstaking ways. | + | 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 |
| - | Robots can learn from experience and can improve their behavior | + | expert interview on “Human-robotics relations: What does the |
| - | over time. In this talk I will give an overview of methods, | + | 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 |
| - | opportunities, and challenges of machine learning in | + | Interaction</i> at <a href="https://airov.at/2026/workshop/TactileRobotics.html">Austrian |
| - | robotics.</blockquote></td></tr><tr valign="top"><td>2020-11-20</td><td>Simon Haller-Seeber and Patrick Lamprecht present a show <i>Explainable AI: A sneak peek into the Black-Box</i> at <a href="https://youtube.com/watch?v=4NVfPwdLnCg&t=17m43s">Science Slam</a>, online.</td></tr><tr valign="top"><td>2020-09-30</td><td>Erwan Renaudo contributes a talk <i>ROSSINI: RobOt kidS deSIgn thiNkIng</i> at <a href="https://www.springer.com/gp/book/9783030674106">Robotics in | + | 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 |
| - | Education 2020</a>, online.</td></tr><tr valign="top"><td>2020-06-22</td><td>Justus Piater gives an invited talk <i>Conditional Neural Movement Primitives</i> at <a href="http://www.gdr-isis.fr/index.php?page=reunion&idreunion=424">GdR | + | Grenzen</i> at SchulleiterInnen-Tagung „IT-Sicherheit und KI-Einsatz in |
| - | ISIS Réunion Apprentissage et Robotique</a>, online. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem4', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem4Abstract" style="display:none">Conditional Neural Movement Primitives (CNMP) constitute | + | 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 |
| - | a novel framework for robot programming by demonstration based on | + | die Welt im Sturm erobert. Wie können wir sie produktiv nutzen? |
| - | Conditional Neural Processes (CNP). Like Bayesian methods such as | + | Wie können wir mit den Herausforderungen umgehen, die durch sie |
| - | Gaussian Processes (GP), CNP learn how target distributions depend | + | entstehen? Ich werde diese Fragen von der technischen Seite her |
| - | on data, and can be conditioned on specific data points to infer | + | angehen und einen Eindruck davon vermitteln, wie diese Systeme |
| - | new target distributions at test time. Unlike GP that are | + | funktionieren. Daraus ergeben sich ein realistisches Verständnis |
| - | expensive to train and scale poorly to high dimensions, CNP are | + | für ihre Möglichkeiten und Grenzen sowie einige grundlegende |
| - | neural networks and are trained by gradient descent. CNMP | + | 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. |
| - | leverage CNP to represent motion trajectories that can be | + | 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 |
| - | conditioned, at test time, on task paramters such as goal | + | 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 |
| - | locations, via points, and/or force readings. Moreover, CNMP are | + | sophistication of robots has been rising with costs falling. Yet, |
| - | conditioned on sensor readings during execution, resulting in | + | the capabilities of AI-enabled robots are not keeping pace. I |
| - | robust, reactive behavior. This talk will present an overview of | + | argue that this is due to a lack of structural understanding by |
| - | how CNMP work and how they can be used in various robot | + | current AI systems. I will discuss several lines of research in my |
| - | applications.</blockquote></td></tr><tr valign="top"><td>2020-06-03</td><td>Justus Piater appears in the media: Wie der Roboter denken lernt.</td></tr><tr valign="top"><td>2020-01-29</td><td>Justus Piater gives an invited talk <i>Digital Science</i> at <a href="https://www.graduateacademy.uni-heidelberg.de/karriere/veranstaltungsreihen.html">Vortragsreihe | + | lab that seek to enable robots to generalize better and learn |
| - | „Primers for Predocs – Strategien für eine erfolgreiche | + | 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, |
| - | Promotion“</a>, Universität Heidelberg. <span class="actions"><a href="javascript:void(0)" onclick="showHide('newsitem7', 'Abstract')">[Abstract]</a></span><blockquote id="newsitem7Abstract" style="display:none">Massive availability of data and computing power are | + | a curious question arises: Where are all the real-world, intelligent robots? |
| - | promoting data-driven methods in all areas of science and | + | While we have mastered the generation of text, images, and video by leveraging vast web-scale datasets, |
| - | technology. I will describe how the University of Innsbruck | + | robotics still faces a fundamental data bottleneck. Reinforcement learning (RL) offers a compelling solution, |
| - | supports this via its new Digital Science Center, and will give a | + | enabling agents to learn autonomously by collecting their own experience. |
| - | flavor of machine learning for data analysis.</blockquote></td></tr><tr valign="top"><td>2020-01-20</td><td>Joanna Chimiak-Opoka, Carina König, and Justus Piater appear in the media: <a href="https://www.uibk.ac.at/newsroom/ergaenzung-digital-science-erfolgreich-gestartet.html.de">Ergänzung Digital Science erfolgreich gestartet – UIBK Newsroom</a>.</td></tr><tr valign="top"><td>2020-01-03</td><td>Justus Piater gives an invited talk <i>Künstliche Intelligenz: Grundlagen, Erfolge, | + | However, the path to autonomy is often blocked by the staggering inefficiency of current RL algorithms, |
| - | Herausforderungen</i> at 47. Tagung des Innsbrucker Kreises von MoraltheologInnen | + | which can require millions of trials to master simple tasks. This talk embarks on a quest to tackle this |
| - | und SozialethikerInnen, Innsbruck.</td></tr><tr valign="top"><td>2019-12-19</td><td>Justus Piater appears in the media: <a href="https://tvthek.orf.at/profile/Tirol-heute/70023/Tirol-heute/14035670/Alexa-Siri-Co-Spione-im-eigenen-Haus/14610743">TV interview by ORF 2 Tirol Heute RedHaus (in German)</a>.</td></tr><tr valign="top"><td>2019-12-12</td><td>Justus Piater gives an invited lecture <i>Too Smart to Be Trusted – Do I Even Want to | + | efficiency problem head-on. I will argue that a crucial step toward unlocking the potential of |
| - | Understand My Robot?</i> at <a href="http://www.trustrobots.eu/">TrustRobots Lecture series | + | robot learning lies in a two-pronged approach: first, by developing statistically efficient |
| - | Trust in Robots</a>, TU Vienna.</td></tr></table></HTML> | + | algorithms that can reuse data by leveraging sound off-policy techniques, and second, |
| + | by designing better action representations for physical, real-world agents. | ||
| + | By making our algorithms more efficient and refining their core hypotheses, | ||
| + | we can accelerate the journey toward real embodied artificial intelligence.</blockquote></td></tr></table></HTML> | ||
| [[news|Older News]] | [[news|Older News]] | ||
| Line 76: | Line 79: | ||
| Austria | Austria | ||
| - | **How to find us:** See the [[http://informatik.uibk.ac.at/how-to-reach-us/|directions]]. | + | **How to find us:** See the [[https://www.uibk.ac.at/informatik/kontakt/anfahrt.html.en|directions]]. |
| **Legal Notice:** See the [[:impressum|Impress and Privacy Notice]]. | **Legal Notice:** See the [[:impressum|Impress and Privacy Notice]]. | ||
start.1607577922.txt.gz · Last modified: 2020/12/10 06:25 by IIS Webadmin