<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0"
	 xmlns:content="http://purl.org/rss/1.0/modules/content/"
	 xmlns:utevent="http://uni-tuebingen.de/ns/event/"
	 xmlns:f="http://typo3.org/ns/TYPO3/CMS/Fluid/ViewHelpers"
	 xmlns:ut="http://typo3.org/ns/Unitue/Project/ViewHelpers"
	 xmlns:n="http://typo3.org/ns/GeorgRinger/News/ViewHelpers">
	<channel>
		<title>Home Wichmann-Lab</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/home/</link><description>Der RSS Feed der Universität Tübingen</description><language>en-EN</language><copyright>Universität Tübingen</copyright><pubDate>Wed, 11 Nov 2026 05:32:53 +0100</pubDate><lastBuildDate>Wed, 11 Nov 2026 05:32:53 +0100</lastBuildDate><item><guid isPermaLink="false">news-138603</guid><pubDate>Wed, 04 Nov 2026 09:32:45 +0100</pubDate><title>New paper published in the Journal of Vision</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138603&amp;cHash=37d438f0bb827528782d5c1d632b0667</link><description>Title: &quot;The spatial area of the normalization pool in early visual processing&quot; by Aqsa Hassan, Felix A. Wichmann, Rafal K. Mantiuk and Heiko H. Schütt</description><content:encoded><![CDATA[<p><strong>Abstract:</strong><br>Precisely measuring the properties of low-level visual processing is of theoretical interest as such properties often generalize to other processing in the brain and have immediate applications for image and video processing. The initial stages of human visual processing split the signal into spatial frequency and orientation-tuned channels, which interact nonlinearly, are typically modeled as divisive normalization. While we know in detail which spatial frequency and orientation channels interact, we know much less about how far the interactions reach across space. In particular, we lack psychophysical data to constrain the spatial area of the normalization pool. To provide such constraints, we set up forced choice masking experiments varying the size of the mask while we keep the spatial frequency and orientation of the mask the same as the targets. The target grating was always a small (0.5 cycles standard deviation) circular Gabor centrally superimposed over 15% and 40% contrast masks. We tested eight different mask sizes for 1, 2, and 4 cycles per degree vertical stimuli in the fovea and at 5° and 10° eccentricities. In the periphery at 5° and 10° eccentricities, contrast thresholds increase with increasing mask size and then saturate as expected, which allows us to estimate the spatial extent of the normalization pool by fitting a divisive normalization model. The fitted Gaussian normalization pool has a standard deviation of 0.65° for 1 cycle per degree gratings and scaled inversely proportional to the spatial frequency of the gratings. Neither the eccentricity nor the mask contrast influenced the size of the normalization pool. In the fovea, however, the size of the mask had no influence on thresholds, even though the mask reliably elevated thresholds. Thus, we are left with an accurate estimate of normalization pool sizes in the periphery, along with the unresolved question of why masking in the fovea shows a different behavior.</p><p>To see the whole article, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/peer-reviewed/#c2388621" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138591</guid><pubDate>Tue, 03 Nov 2026 16:23:08 +0100</pubDate><title>Felix Wichmann will give a lecture at &quot;VISION WEEK: brain, mind, and machine&quot; </title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138591&amp;cHash=af395e52107eb96ad107dc1528a3ec7b</link><description>at the Paris Université Sciences &amp; Lettres (PSL) in early March 2027. </description><content:encoded><![CDATA[<p>Please visit this website to get more information about the university: <a href="https://psl.eu/en" target="_blank" rel="noreferrer">https://psl.eu/en</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-137550</guid><pubDate>Thu, 15 Oct 2026 14:22:14 +0200</pubDate><title>At the beginning of May 2027 Felix Wichmann will be giving a keynote at the &quot;Lucerne Summer University: Ethics in a Global Context (LSUE) 2027&quot; </title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=137550&amp;cHash=cb565349e2ee664eaaa6d9cc76c963d1</link><description>in Lucerne, Switzerland. </description><content:encoded><![CDATA[<p>Link: <a href="https://www.unilu.ch/fakultaeten/tf/institute/institut-fuer-sozialethik-ise/lucerne-summer-university-ethics-in-a-global-context-lsue/" target="_blank" rel="noreferrer">https://www.unilu.ch/fakultaeten/tf/institute/institut-fuer-sozialethik-ise/lucerne-summer-university-ethics-in-a-global-context-lsue/</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-137547</guid><pubDate>Thu, 15 Oct 2026 14:20:03 +0200</pubDate><title>Felix Wichmann will give a keynote at the &quot;Measuring Behaviour&quot; 2027 Conference in Düsseldorf</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=137547&amp;cHash=cad2f5f3d4a5e5c743bd06910dc1b4e2</link><description>at the end of April 2027</description><content:encoded><![CDATA[<p>Please see <a href="https://www.measuringbehavior.org/" target="_blank" rel="noreferrer">https://www.measuringbehavior.org/</a> for details</p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-137544</guid><pubDate>Thu, 15 Oct 2026 14:04:03 +0200</pubDate><title>We have a new PhD student: Katharina Holzhey</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=137544&amp;cHash=a1b5c8b0d8b37eb59d2979e4ec31f6c3</link><description>She has joined our lab in October 2026 and we are very glad to welcome her.</description><content:encoded><![CDATA[<p>Katharina will partly be funded by the <a href="https://uni-tuebingen.de/en/research/core-research/cluster-of-excellence-machine-learning/home/" target="_blank" class="ut-link ut-nav-area__link ut-nav-area__link--current" title="Cluster of Excellence – Machine Learning for Science">Cluster of Excellence – Machine Learning for Science</a>.</p><p>Please have a look at our <a href="/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/people/members/#c2613177">members page</a> to read more about her research plans.</p><p>Welcome, Katharina!</p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138219</guid><pubDate>Thu, 24 Sep 2026 11:50:00 +0200</pubDate><title>New article published at NeurIPS 2026</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138219&amp;cHash=88bef9a862b9082dfe820398a4fb4488</link><description>Title: &quot;Parameter Symmetries Determine Representational Geometry in Overparameterized Nonlinear Networks&quot; by Marvin Theiss, Lukas Braun, Andrew M. Saxe and Erin Grant</description><content:encoded><![CDATA[<p><strong>Abstract:</strong><br>Representations are routinely used across machine learning, psychology, and neuroscience to draw inferences about the computations of biological and artificial systems. Such inferences presume that the geometry of these representations reflects the computation being performed. For artificial neural networks, however, the extent to which the function a network computes constrains this geometry remains unclear. One key obstacle is that these networks admit parameter symmetries: changes in parameterization that preserve function exactly while reshaping representational geometry. Here, we show that a broad class of parameter symmetries acts on representations through just three primitive feature transformations: addition, duplication, and scaling. This feature-level characterization yields a closed-form decomposition of representational geometry into essential and auxiliary components, revealing how representational degeneracy can exist without and grows with overparameterization even when function is held fixed. Finally, we establish conditions under which implementation-level selection rules resolve this degeneracy, yielding identifiable geometries in which features are weighted according to their contributions to the network's function. Together, our results delineate when representational geometry can support inferences about computation, and when it cannot.</p><p>To see the whole article, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/peer-reviewed/#c2388621" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138216</guid><pubDate>Fri, 10 Jul 2026 11:48:00 +0200</pubDate><title>Marvin receives Best Paper Award at ICML 2026 Workshop</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138216&amp;cHash=7b3d78c3d9b66b7db4739a58ff8155d0</link><description>His paper &quot;Parameter symmetries determine representational geometry in overparameterized nonlinear networks&quot; co-authored by Lukas Braun, Andrew M. Saxe and Erin Grant has been recognised with the Best Paper Award at the ICML 2026 Workshop on Weight-Space Symmetries: From Foundations to Practical Applications.</description><content:encoded><![CDATA[<p>Congratulations, Marvin!</p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-133392</guid><pubDate>Fri, 03 Jul 2026 13:46:15 +0200</pubDate><title>New abstract accepted as a poster at ECVP 2026, Bournemouth, GB</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=133392&amp;cHash=f79e56450a036b8f91faaac789cf0948</link><description>Title: &quot;Examining spatial and temporal processing of low spatial frequency sine and square gratings&quot; by M. Jannati and F. A. Wichmann</description><content:encoded><![CDATA[<p>To see the whole abstract, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/conference-abstracts/#c2445306" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-137568</guid><pubDate>Fri, 03 Jul 2026 13:46:15 +0200</pubDate><title>New abstract accepted as a poster at ECVP 2026, Bournemouth, GB</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=137568&amp;cHash=ed2deefef179c7c5383c0ae8beb6da22</link><description>Title: &quot;Improving a fast and modular image-computable psychophysical spatial vision model&quot; by J. Reichert and F. A. Wichmann</description><content:encoded><![CDATA[<p>To see the whole abstract, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/conference-abstracts/#c2445306" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-133389</guid><pubDate>Fri, 03 Jul 2026 13:43:32 +0200</pubDate><title>New article accepted as a talk at CogSci 2026, Rio de Janeiro, BR</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=133389&amp;cHash=69113b10f728d7f637234c9d595bac51</link><description>Title: &quot;Mapping the Folk Concept of Health&quot; by Lukas S. Huber, Kevin Reuter and Somogy Varga</description><content:encoded><![CDATA[<p>To see the whole article, please visit our <a href="/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/peer-reviewed/#c2388621">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138213</guid><pubDate>Wed, 17 Jun 2026 11:43:00 +0200</pubDate><title>Oral presentation at ICML 2026 Workshop on Weight-Space Symmetries</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138213&amp;cHash=02fc14585d255ca3c3313e33faa195f7</link><description>TItle: &quot;Parameter symmetries determine representational geometry in overparameterized nonlinear networks&quot; by Marvin Theiss, Lukas Braun, Andrew M. Saxe and Erin Grant</description><content:encoded><![CDATA[<p><strong>Abstract:</strong><br>Representations are routinely used in machine learning, psychology, and neuroscience to probe the computations of biological and artificial systems. Yet it remains unclear to what extent computation constrains representation in artificial neural networks. One key obstacle is that these networks admit parameter symmetries: transformations of the parameters that preserve function exactly while reshaping representational geometry. Here we show that known parameter symmetries act on representations through just three primitives: addition, duplication, and scaling. This yields a closed-form descriptor of representational geometry as a sum of task-linked features and symmetry-induced noise. This decomposition further provides analytic bounds on representational similarity under parameter symmetries, revealing when functionally equivalent networks can become arbitrarily dissimilar. Finally, we identify privileged representational geometries, which weight features by their computational importance and recover a stable link between representation and computation. Overall, our results delineate when representation can support inferences about computation, and when it cannot.</p><p>To see the whole article, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/peer-reviewed/#c2388621" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138210</guid><pubDate>Tue, 02 Jun 2026 11:40:00 +0200</pubDate><title>New abstract accepted as a poster at CCN 2026, New York, NY, USA</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138210&amp;cHash=e73c37485af934273937a96706099727</link><description>Title: &quot;Representation can dissociate from function and behaviour in task-optimised vision networks&quot; by Marvin Theiss, Ben Fausten and Felix A. Wichmann</description><content:encoded><![CDATA[<p>To see the whole abstract, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/conference-abstracts/#c2445306" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138204</guid><pubDate>Tue, 02 Jun 2026 11:37:00 +0200</pubDate><title>New abstract accepted as a poster at CCN 2026, New York, NY, USA</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138204&amp;cHash=139f2b6cfd2a7d4ec8bb2b63c8abbd72</link><description>Title: &quot;Parameter symmetries determine representational geometry in overparameterized nonlinear networks&quot; by Marvin Theiss, Lukas Braun, Andrew M. Saxe and Erin Grant</description><content:encoded><![CDATA[<p>To see the whole abstract, please visit our <a href="https://uni-tuebingen.de/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/publications/conference-abstracts/#c2445306" target="_blank">publications page.</a></p>]]></content:encoded><category>NeuroInfo</category></item><item><guid isPermaLink="false">news-138192</guid><pubDate>Mon, 04 May 2026 11:32:00 +0200</pubDate><title>Marvin Theiß to co-organise keynote &amp; tutorial session at CCN 2026, New York, NY, USA</title><link>https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/neuronale-informationsverarbeitung/news/newsfullview-aktuell/?tx_news_pi1%5Baction%5D=detail&amp;tx_news_pi1%5Bcontroller%5D=News&amp;tx_news_pi1%5Bnews%5D=138192&amp;cHash=a1ff98d900c46d9f12fc6f94bc7ba579</link><description>Title: &quot;Adaptive, degenerate, and yet comparable? Rethinking representational comparisons&quot; organised by Erin Grant, Lukas Braun, Marvin Theiss and Eleanor Holton</description><content:encoded><![CDATA[<p><strong>Abstract:</strong><br>Comparing patterns of neural activity within and between biological and artificial systems has become a core methodology in neuroscience, psychology, and machine learning. Motivated by the goal of advancing our understanding of neural computation, representational comparisons are used to infer shared or divergent computations in minds and machines. However, neural network theory challenges the assumption that representational similarity generally provides a measure of computational similarity and vice versa. This tension arises from two defining features of biological and artificial neural networks: adaptivity and degeneracy. While internal representations are adapted to task demands and objectives, many distinct network configurations with distinct internal representations can implement the same behavior. Adaptivity therefore motivates the use of artificial neural networks as normative models of representation learning, whereas degeneracy challenges the assumption that comparing patterns of neural activity is, by itself, a well-posed basis for inferring and comparing computation. Starting from simple two-layer linear networks and extending to deep nonlinear networks, we show how function and neural representation are dissociable and reveal theoretically grounded conditions under which representational comparisons become scientifically meaningful. We argue that in light of degeneracy, a central object of study in cognitive computational neuroscience should be representational plurality: the existence of distinct neural geometries that support identical behavior but differ in their downstream computational affordances. We show how representational plurality can be identified and investigated through theoretically motivated task design and behavioral readouts. Overall, the keynote uses neural network theory to delineate when representational geometry can and cannot support inference about computation, and argues for going beyond naive representational comparisons towards the study of representational formation, differentiation, and adaptation.</p><p>For more information, visit the <a href="https://eringrant.github.io/ccn-2026-kt/" target="_blank" rel="noreferrer">project website</a>.</p>]]></content:encoded><category>NeuroInfo</category></item>
	</channel>
</rss>

