Intelligent and Interactive Systems

User Tools

Site Tools


start

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
start [2020/11/26 06:25]
IIS Webadmin
start [2022/05/20 06:25] (current)
IIS Webadmin
Line 11: Line 11:
  
 ===== Working With Us ===== ===== Working With Us =====
 +  * We are hiring [[jobs|two doctoral students]] in machine learning for computer vision or robotics.
   * 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]]+ <​html><​!-- 
 + * [[https://​ellis.eu/​|ELLIS]] PhD Program: [[https://​ellis.eu/​news/​ellis-phd-program-call-for-applications-deadline-november-15-2021|Call for Applications]] 
 +--></​html>​
  
 <​html><​!-- <​html><​!--
Line 29: Line 32:
 ===== 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>​2022-06-02</​td><​td>​Erwan Renaudo contributes a talk <i>Deep Learning ​for Fast Segmentation of E-waste Devices’  
-    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 +    Inner Parts in a Recycling Scenario</i> at <a href="​https://​icprai2022.sciencesconf.org/">​ICPRAI 2022</​a>, ​Paris/online. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem0', '​Abstract'​)">​[Abstract]</​a></​span><​blockquote id="newsitem0Abstract" style="​display:​none">​Recycling obsolete electronic devices (E-waste) is a 
-    ​learning capabilities and flexibility This will enable them to +    ​dangerous task for human workersAutomated E-waste recycling ​ 
-    ​act purposefully in unstructured environments and to react to +    ​is an area of great interest but challenging for current robotic ​ 
-    ​unforeseen events ​People can teach them intuitively to perform +    ​applicationsWe focus on the problem of segmenting inner parts  
-    ​tasks instead ​of having to program them in painstaking ways+    of E-waste devices into manipulable elementsFirst, we extend ​ 
-    ​Robots can learn from experience ​and can improve their behavior +    ​a dataset of hard-drive disk (HDD) components with labelled  
-    ​over time In this talk I will give an overview ​of methods+    occluded ​and non-occluded points of view of the parts, in order  
-    ​opportunities, ​and challenges ​of machine ​learning ​in +    ​to increase the diversity and the quality of the learning data  
-    ​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: sneak peek into the Black-Box</i> at <a href="​https://​youtube.com/watch?v=4NVfPwdLnCg&​amp;​t=17m43s">Science Slam</​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&​amp;​idreunion=424">GdR +    with different anglesWe then perform ​an extensive evaluation  
-    ​ISIS Réunion Apprentissage et Robotique</​a>, ​online. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem3', '​Abstract'​)">​[Abstract]</​a></​span><​blockquote id="newsitem3Abstract" style="​display:​none">​Conditional Neural Movement Primitives (CNMP) constitute +    with three different state-of-the-art models, namely CenterMask
-    ​novel framework for robot programming by demonstration based on +    ​BlendMask ​and SOLOv2 (including variants) and two types of  
-    Conditional Neural Processes (CNP). ​ Like Bayesian methods such as +    metrics: the average precision as well as the frame rate. Our  
-    Gaussian Processes (GP), CNP learn how target distributions depend +    results show that instance segmentation using state-of-the-art  
-    on data, and can be conditioned on specific data points to infer +    deep learning ​methods can precisely detect complex shapes along  
-    ​new target distributions at test time.  Unlike GP that are +    ​with their boundaries, as well as being suited for fast tracking of 
-    expensive ​to train and scale poorly to high dimensions, CNP are +    parts in a robotic recycling system(Rojas et  al. 2022) 
-    ​neural networks and are trained ​by gradient descent. ​ CNMP +    ​</​blockquote></​td></​tr><​tr valign="​top"><​td>​2022-05-17</​td><​td>​Erwan Renaudo gives an invited talk <​i>​A ​Brief tour of autonomous robots'​ &​quot;​Cognition&​quot;​</i> at <a href="​https://​www.philosophie.uni-konstanz.de/ag-mueller/​aktuelles/​meldungsdetails-ag-mueller/​2022/​5/​17/​event/​46426-Creating-Agency-and-Cogn/​tx_cal_phpicalendar/">​Creating Agency and Cognition in Automated Systems: What can we learn from the OctopusInternational Hybrid Workshop</​a>,​ Innsbruck. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem1',​ '​Abstract'​)">[Abstract]</a></​span><​blockquote id="​newsitem1Abstract"​ style="​display:​none">​In the design of &​quot;​Cognitive Robots&​quot;​ software architectures 
-    ​leverage CNP to represent motion trajectories that can be +    researchers had to face many organizational problems due to  
-    ​conditioned,​ at test time, on task paramters such as goal +    the particular nature of robotsNot only should they manage  
-    locations, via points, ​and/or force readings. ​ MoreoverCNMP are +    to achieve complex tasks with long-terms goals, but they must  
-    ​conditioned on sensor readings during executionresulting in +    also be reactive to their surroundings to preserve their environment. 
-    robust, reactive behavior. ​ This talk will present an overview of +    We will go over several layered control paradigms developed over 
-    how CNMP work and how they can be used in various robot +    the history of robotics and see how these can relate to the  
-    ​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 +    knowledge about the Octopus'​ brain organization,​ and discuss  
-    ​„Primers for Predocs – Strategien für eine erfolgreiche +    how the Octopus itself can inspire cognitive robotic research. 
-    Promotion“</​a>, ​Universität Heidelberg. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem6', '​Abstract'​)">​[Abstract]</​a></​span><​blockquote id="newsitem6Abstract" style="​display:​none">​Massive availability of data and computing power are +    </​blockquote>​</​td></​tr><​tr valign="​top"><​td>​2022-05-11</​td><​td>​Justus Piater gives an invited talk <i>Robots That Learn Like Humans?</i> at <a href="https://pintofscience.at/events/​innsbruck">Pint of 
-    ​promoting data-driven ​methods in all areas of science ​and +    ​Science</​a>, ​Innsbruck. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem2', '​Abstract'​)">​[Abstract]</​a></​span><​blockquote id="newsitem2Abstract" style="​display:​none">​Wouldn'​t it be great to have helper ​robot that we can 
-    ​technology I will describe how the University of Innsbruck +    ​task with our annoying chores? How would we teach the robot to 
-    ​supports ​this via its new Digital Science Center, and will give a +    ​perform these tasks? Can we build such robots ​by equipping them 
-    ​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, +    ​with artificial intelligence?​ We will discuss how human and 
-    ​Herausforderungen</i> at 47Tagung des Innsbrucker Kreises von MoraltheologInnen +    ​machine learning differ ​and how this impacts what skills
-    ​und SozialethikerInnenInnsbruck.</​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 +    ​including for communicationour helper robot will be able to 
-    Understand My Robot?</i> at <a href="http://www.trustrobots.eu/">TrustRobots Lecture series +    ​learn.</​blockquote></​td></​tr><​tr valign="​top"><​td>​2022-02-19</​td><​td>​Justus Piater gives an invited talk <i>Learning as a Creative Process in Humans and 
-    Trust in Robots</​a>, ​TU Vienna.</​td></​tr><​tr valign="​top"><​td>​2019-12-05</​td><​td>​IIS guest <span style="​font-weight:​bold">​Heiko Neumann</​span>,​ University of Ulm,  gives an invited ​colloquium ​<i>Biologically inspired visual-auditory processing – from +    Machines</i> at <a href="​https://​www.philosophie.uni-konstanz.de/ag-mueller/​online-workshop-on-agency-life-and-creativity/">Workshop 
-    brain-like computation ​to neuromorphic algorithms</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('​newsitem11',​ '​Abstract'​)">[Abstract]</​a></​span><blockquote id="newsitem11Abstract"​ style="​display:none">A fundamental task of sensory processing is to detect +    ​on Agency, Life, and Creativity”</​a>, ​online. <span class="​actions"><​a href="​javascript:​void(0)"​ onclick="​showHide('​newsitem3', '​Abstract'​)">​[Abstract]</​a></​span><​blockquote id="newsitem3Abstract" style="​display:​none">​Humans constantly engage in creative activity such as 
-    ​and integrate feature items to group them into perceptual units +    finding explanations,​ improving technology, or solving novel 
-    segregating them from other objects and the background. A +    problems in everyday life. Even imitating other humans, e.g. while 
-    framework is discussed which explains how perceptual grouping at +    learning a new skill, is creative in that the imitator has to 
-    early as well as higher-level cognitive stages may be implemented +    infer the conceptual structure underlying the observed 
-    in cortex. Different grouping mechanisms are implemented which are +    manifestation ​and to regenerate the latter from the former. In 
-    attuned to basic features and feature combinations and mainly +    ​contrast, current AI-enabled agents are unable to generate and 
-    evaluated along the forward sweep of stimulus processing. However, +    draw upon such conceptual structures. Their learning is mostly 
-    due to limitations of local feature detection mechanisms and +    ​driven ​by statistics, ​and any ability to produce novelty is 
-    inherent ambiguities,​ top-down feedback is required to deliver +    ​largely limited to trial and error. I will try to characterize 
-    contextual information helping to disambiguate initial +    this fundamental difference between learning in human and 
-    measurements. Feedback of contextual information is demonstrated +    artificial agents, and will discuss possible technological 
-    to improve object recognition performance,​ stabilize learning of +    ​approaches that might help close this gap.</​blockquote></​td></​tr><​tr valign="​top"><​td>​2021-11-24</​td><​td>​Justus Piater ​is a panelist at a public discussion on <a href="​https://​www.linkedin.com/posts/elsa-innsbruck_letzte-woche-fand-zum-elsa-day-am-24112021-activity-6871397038093869057-16-7">Privacy 
-    object categories, and integrate multi-sensory representations. +    in the Digital ​Age</a>, online(Organized by ELSA Innsbruck)</​td></​tr><​tr valign="​top"><​td>​2021-10-22</​td><​td>​Justus Piater gives an invited talk <i>Robotik und KI in der Medizin – nur 
- +    ​Bedarfserweckung?​</i> at <a href="​https://​medical-update-hall.com/​veranstaltungen/">​Medical 
-    The canonical principles of neural computation define ​set of +    ​Update Hall 2021 – Fit für das neue Jahrzehnt</​a>​UMIT, Hall in Tirol.</​td></​tr><​tr valign="​top"><​td>​2021-08-26</​td><​td>​Justus Piater ​contributes a talk <​i>​Digital Science at the University of Innsbruck</​i>​ at Aurora Research Conference Digital Society &amp; Global Citizenship,​ Vrije Universiteit Amsterdam. <span class="​actions">​<a href="javascript:void(0)"​ onclick="​showHide('​newsitem6',​ '​Abstract'​)">​[Abstract]<​/a></span><​blockquote id="​newsitem6Abstract"​ style="​display:​none">​Growing data processing capacities and progress in 
-    core operations to implement above-mentioned mechanisms of +    analytical methods and artificial intelligence are motivating 
-    perceptual and cognitive inference. These operations can be +    entire branches of science to raise new questions and to develop 
-    mappedin a simplified form, onto neuromorphic platforms to +    new approachesThis digital transformation requires progressive,​ 
-    emulate brain-like computationIt is demonstrated that an +    interdisciplinary synergies between the computational sciences and 
-    architecture composed of canonical circuit mechanisms can be +    all other scientific disciplines The Digital Science Center 
-    mapped onto neuromorphic chip technology facilitating low-energy +    (DiSC) ​at the University of Innsbruck meets this challenge ​by 
-    non-von Neumann computation.</​blockquote>​</​td></​tr></​table></​HTML>​+    integrating and promoting digitalisation of scientific research. 
 +    In this talk I will present some examples of interdisciplinary 
 +    research by DiSC staff researchers who are experts ​in specific 
 +    aspects of digital science.</blockquote></​td></​tr><​tr valign="​top"><​td>​2021-08-24</​td><​td>​Justus Piater ​contributes a lecture <i>Picking and Placing</i> at <a href="https://www.uibk.ac.at/informatik/​forschung/​gmar-robotics-school-2021.html">2021 GMAR Summer School</​a>, ​Villa Blanka, Innsbruck.</​td></​tr><​tr valign="​top"><​td>​2021-07-01</​td><​td>​Justus Piater and Matteo Saveriano give an invited ​talk <i>Making Robots Learn to Perceive and Act with Understanding</i> at <a href="​https://​webservices.scientificnet.org/rest/entries/api/v1/​blobs/​200894">Workshop 
 +    and Open Lab on Field Robotics - Interdisciplinary aspects of 
 +    robotics and its applications in outdoor scenarios</a>, NOI Techpark, Bolzano, Italy.</td></​tr><​tr valign="top"><​td>​2021-05-26</td><​td>​Justus Piater contributes ​talk <i>Intelligente und interaktive Systeme</iat <a href="http://​www.gmar.at/​aktuell/​">GMAR Robotics Science 
 +    ​Talks</a>online.</​td></​tr></​table></​HTML>​
  
 [[news|Older News]] [[news|Older News]]
start.1606368321.txt.gz · Last modified: 2020/11/26 06:25 by IIS Webadmin