June 2024 - Comp-neuro - lists.cnsorg.org

World wide VVTNS series: Wednesday, June 19 at 11am (EDT), Yasaman Bahri , Google DeepMind | Learning and prediction in artificial deep neural networks: scaling, data manifolds, and universality
by David Hansel June 16, 2024

June 16, 2024

[image: VVTNS.png] https://www.wwtns.online - on twitter: wwtns@TheoreticalWide You are cordially invited to the lecture given by Yasaman Bahri Google DeepMind on the topic of Learning and prediction in artificial deep neural networks: scaling, data manifolds, and universality The lecture will be held on zoom on *Wednesday* *June 19, 2024*, at *11:00 am EDT * To receive the link: https://www.wwtns.online/register-page *Abstract: *Developing scientifically-grounded theories for representation learning and generalization in artificial deep neural networks remains a grand challenge of fundamental interest to theoretical neuroscience and machine learning. I will discuss our work on one facet of this challenge — namely understanding generalization or “scaling laws” in learned neural networks as a function of basic control variables. I’ll discuss a taxonomy we develop that classifies different regimes of scaling behavior. We identify regimes where generalization exhibits universal scaling behavior and others where it can be traced back to properties of the data and neural architecture. The theoretical analysis is enabled by leveraging exactly solvable models of deep neural networks that arise naturally in the limit of large hidden layers. Along the way, I’ll also discuss our work on these theoretical models, which have been a useful starting point for theoretical descriptions of neural network dynamics. Finally, I’ll discuss our findings connecting generalization in neural networks to properties of the learned data manifold. I’ll close by discussing future directions and new hypotheses that emerge from our findings *About VVTNS : Created as the World Wide Neuroscience Seminar (WWTNS) in November 2020 and renamed in homage to Carl van Vreeswijk in Memoriam (April 20, 2022), its aim is to be a platform to exchange ideas among theoreticians. Speakers have the occasion to talk about theoretical aspects of their work which cannot be discussed in a setting where the majority of the audience consists of experimentalists. The seminars, **held on Wednesdays at 11 am ET,** are 45-50 min long followed by a discussion. The talks are recorded with authorization of the speaker and are available to everybody on our YouTube channel.* ᐧ ᐧ ᐧ ᐧ ᐧ

1 0

2024 SPARC FAIR Codeathon: apply now and choose a project
by Elisabetta Iavarone June 14, 2024

June 14, 2024

Dear all, The Common Fund SPARC program, with funding from the National Institutes of Health (NIH), has a three-day in-the-cloud Codeathon that may be of interest for the comp-neuro community. The fourth annual virtual SPARC Codeathon will take place *online **August 10-12*. This global prize competition is currently recruiting participants to make SPARC data and resources <https://sparc.science/> more Findable, Interoperable, Accessible, and Reusable (FAIR). Any interested participant can join projects spanning data visualization, simulation tool development, python workflow creation, and more to compete for $50,000 in prizes. Follow this link to learn more and sign up today! https://sparc.science/news-and-events/events/2024-sparc-fair-codeathon Elisabetta Iavarone, On behalf of the SPARC FAIR Codeathon team

1 0

Open position at Goettingen University - PhD student (f/m/d) - SPP 2205
by von Behren, Christina June 14, 2024

June 14, 2024

The Department of Data-driven Analysis of Biological Networks at the Georg-August-Universität of Göttingen invites applications for a position as PhD student (f/m/d)<https://www.uni-goettingen.de/de/644546.html?details=2851/> This position is available starting from November 1st, 2024. It is a part time position with a limited contract of 3 years. Salary and working hours will be according to the German grade 13 TV-L (65%, currently 25,87 hours per week). The position will be associated with the group of Prof. Dr. Michael Wibral, Göttingen Campus Institute for Dynamics of Biological Networks (CIDBN)<https://www.uni-goettingen.de/de/608362.html/>. We are recruiting a PhD student for the interdisciplinary research project "Evolutionary Convergence of Hierarchical Information Processing" funded by the German Research Foundation (DFG) within the research initiative SPP2205 "Evolutionary optimization of neuronal processing<https://www.uni-goettingen.de/de/618311.html>. The project aims to establish the common information theoretic principles of computation in hierarchical neural systems that are conserved across species (in particular mammals and birds). Specifically, we will use the analysis of information theoretic timescales [1], the correlation of information storage and transfer [2] and the analysis of unique, redundant and synergistic information [3,4] in order to take a comparative approach to neural computation in the primate and the avian brain. The project will be pursued in close collaboration with the research groups of Viola Priesemann at the Max Planck Institute for Dynamics and Self-Organization in Göttingen and Onur Güntürkün, Jonas Rose and Roland Pusch at the Ruhr University of Bochum. We are looking for candidates with an excellent Master degree in physics, neuroscience, cognitive science, computer science or a related field who are committed to pursuing a career in research. The successful applicant will develop novel information theoretic models and measures of neural computation and apply them to electrophysiological recordings to analyze predictive hierarchical information processing in the primate and the avian brain. The ideal candidate should possess exceptional analytical skills, coupled with a proficiency in coding in a language that is pertinent to data analysis. We offer the opportunity to become a member of a vibrant community dedicated to innovative research and teaching at the frontier of the exact and biological sciences, and to connect to the latest developments in information theory, in particular partial information decomposition. The Göttingen Campus is a leading center of biophysics and neuroscience in Europe hosting numerous internationally renowned research institutions, including the University and its Medical Center, the three Max Planck Institutes in the Life Sciences, the European Neuroscience Institute, the German Primate Center, and the Bernstein Center for Computational Neuroscience (BCCN) Göttingen. The University of Göttingen is an equal opportunities employer and places particular emphasis on fostering career opportunities for women. Qualified women are therefore strongly encouraged to apply in fields in which they are underrepresented. The university has committed itself to being a family-friendly institution and supports their employees in balancing work and family life. The university is particularly committed to the professional participation of severely disabled employees and therefore welcomes applications from severely disabled persons. In the case of equal qualifications, preference is given to applications from people with severe disabilities. A disability or equality is to be included in the application in order to protect the interests. Please submit your application in digital form only and send all relevant documents in one single PDF-document, including a letter of interest, your curriculum vitae with a list of publications, copies of your certificates and contact information of at least two references to Wibral_lab(a)uni-goettingen.de by August 1st, 2024. If you have any questions, please contact Prof. Dr. Michael Wibral via E-Mail: michael.wibral(a)uni-goettingen.de Please note: With submission of your application, you accept the processing of your applicant data in terms of data-protection law. Further information on the legal basis and data usage is provided in the Hinweisblatt zur Datenschutzgrundverordnung (DSGVO) https://www.uni-goettingen.de/hinweisdsgvo [1] Rudelt et al. (2021) Embedding optimization reveals long-lasting history dependence in neural spiking activity. PLoS Comp Biol 17(6):e1008927 [2] Wollstadt et al. (2023) Information-theoretic analyses of neural data to minimize the effect of researchers' assumptions in predictive coding studies. PLOS Comp Biol 19 (11) e1011567 [3] Gutknecht et al. (2021) Bits and pieces: Understanding information decomposition from part-whole relationships and formal logic. Proc. Royal Soc. A 477 (2251), 20210110 [4] Ehrlich et al. (2023) A measure of the complexity of neural representations based on partial information decomposition, Transactions on Machine Learning Research 5 ------ Goettingen Campus Institute for Dynamics of Biological Networks Bernstein Center for Computational Neuroscience Goettingen University of Goettingen Heinrich-Dueker-Weg 12 37073 Goettingen Germany

1 0

11th rTAIM Online Seminar | Walter Sinnott-Armstrong (Duke Uni.) | 26 June 2024 - 3pm Lisbon Time Zone | Zoom
by steven gouveia June 13, 2024

June 13, 2024

Dear All, Following the 10th rTAIM Online Seminar (available here: https://www.youtube.com/watch?v=ZsRMi76gv9c) we are happy to announce the 11th rTAIM Online Seminar with the participation of Professor Walter Sinnott-Armstrong (Duke University, USA), on the 26th June 2024, 15h-16h30 Lisbon Time Zone, via Zoom. Title: The Moral AI and How We Get There Abstract: Artificial intelligence (AI) is beginning to be used for many life-changing decisions in medicine. Critics object that using AI in these areas is too likely to lead to harm, unfairness, and other moral wrongs. In response, I will argue that these decisions can be made safer and more ethical by building human moral values into the AI decisionmaker. But how can we do that? I will discuss problems for some proposed ways to build morality into AI from the top down and from the bottom up. Then I will explain our lab’s novel hybrid alternative, which surveys human moral judgments and then corrects for ignorance, confusion, and partiality. Because our approach is based on idealized observer theories in ethics, it minimizes substantive assumptions about what is morally right or wrong, and it can be used in a wide variety of contexts. I will report initial empirical results using our method and discuss potential applications to kidney allocation and dementia. Short bio: Walter Sinnott-Armstrong is Chauncey Stillman Professor of Practical Ethics in the Department of Philosophy and the Kenan Institute for Ethics at Duke University. His most recent book is Moral AI and How We Can Get There, with Jana Schaich Borg and Vincent Conitzer in 2024. Zoom: Link: https://videoconf-colibri.zoom.us/j/98415196784?pwd=a3dHbmF4VzlTaVczWTB6QzV… Meeting ID: 984 1519 6784 Passcode: 765321 More info: https://trustaimedicine.weebly.com/<https://trustaimedicine.weebly.com/rtaim-seminars.html>rtaim<https://trustaimedicine.weebly.com/rtaim-seminars.html>-seminars.html<https://trustaimedicine.weebly.com/rtaim-seminars.html> We are also happy to announce that the Call for Abstract for the 3rd International Conference on the Ethics of Artificial Intelligence that will happen in the São Miguel Islands in the Azores (Portugal), between 19-25 September 2024 (online and in-person), is OPEN for submissions (notifications of acceptance/rejectance) are given in 7-10 days). You can find the information related to the 3ICEAI here: https://trustaimedicine.weebly.com/3iceai.html Sincerely, Organizer

1 0

PhD position at Bernstein Center Berlin & Humboldt University (Physics Dep.) available: Universal description of stochastic oscillators
by Benjamin Lindner June 13, 2024

June 13, 2024

*Universal description of stochastic oscillators - higher dimensional examples, extraction of the mapping from data, and networks of oscillators * Within DFG-funded project, I am looking for a PhD candidate to work on the theoretical description of stochastic oscillations.Specifically, we want to apply and generalize a recently developed universal description of stochastic oscillators (Perez-Cervera et al. PNAS, 2023) in three respects: (i) it should be applied to oscillators that have higher dimensions than two going beyond what has been studied before; (ii) the nonlinear transformation at the core of the procedure should be extracted from (simulation or experimental) data; (iii) the theory for coupled stochastic oscillators will be more thoroughly grounded and, moreover, extended to the case of networks of stochastic oscillators (more than two oscillators, the case previously investigated). These explorations and extensions of the framework have potential applications for the theoretical description of many systems outside of thermodynamic equilibrium, in particular in biology and neuroscience. The successful candidate should have a degree in physics, mathematics or computational neuroscience (a background in neurobiology is desirable but not obligatory), programming skills (C++, Python, LaTeX, Linux), an excellent command of the English language, good communication skills, team spirit, and, last but not least, great enthusiasm for interdisciplinary projects. Funding is provided for three years, starting in the *fall 2024*. For details on the doctoral examination process at the Physics Department of Humboldt University Berlin, see https://fakultaeten.hu-berlin.de/en/mnf/wisskar/promotionen/zula. Applications, including a letter of motivation, a CV, and a list of three potential referees should be sent by email to   me benjamin.lindner(a)physik.hu-berlin.de  (cc to officebl(a)bccn-berlin.de) stating the reference number *DR/084/24*. The deadline for applications is *July 24, 2024*, however, later applications might also be considered. Kind regards, Benjamin Lindner -- Benjamin Lindner Professor for Theory of Complex Systems and Neurophysics BCCN Berlin Philippstr. 13, Haus 2, 10115 Berlin Room: 1.17, phone: 0049(0)302093 82492 (new number!!) Physics Dep. Humboldt University Berlin Newtonstr. 15 12489 Berlin Room: 3.412, phone: 0049(0)302093 7934 http://people.physik.hu-berlin.de/~lindner/index.html -----------------------------------------------------

1 0

Postdoc AI machine learning New York
by Sivan Kinreich June 12, 2024

June 12, 2024

Please forward this to those who may be interested: Dr. Kinreich's group at SUNY Downstate Medical Center, NY ( https://www.pmbd2.org/) investigates the structural and functional neural networks (fMRI, EEG), genetics, and other phenotypes associated with the risk and resilience factors underlying the development of brain diseases. We use advanced machine learning algorithms and seek a skilled, motivated, and enthusiastic individual with a Master's degree / postdoctoral fellow to join the lab. The position: Education: Candidates must have a Master's degree or a Ph.D. degree (or towards completion) in computational neuroscience, biology, computer science, data science, or a related discipline. Computer and General Skills: Knowledge of experimental statistics and excellent programming skills are required (Python, Matlab). High-level expertise in neuroimaging analyses is a plus. Previous experience in machine learning analysis and computational modeling is a plus. Takes initiative. Strong communications skills and research productivity, as well as the ability to work effectively in a research team. Duties and Responsibilities: • Managing and analyzing big data, including neuroimaging, genetics, and clinical data • Writing machine learning programs in Python and Matlab to identify patterns in data • Writing manuscripts • Presenting findings at national conferences and local seminars/meetings This job is available immediately for 2 years. Job is open until filled. Salary will be commensurate with experience and qualifications according to NIH scales. Candidates should send their CV, representative publications, and contact information of two references to Sivan Kinreich, email: sivan.kinreich(a)downstate.edu Sivan Kinreich, Ph.D. Department of Psychiatry Prediction Modeling and Biomarker Discovery of Brain Disorders lab (PMBD2) Henri Begleiter Neurodynamics Laboratory SUNY Downstate Medical Center 450 Clarkson Ave. Brooklyn, NY 11203 Tel: 718-270-3322 646-775-8509 Email: sivan.kinreich(a)downstate.edu sivankinreich(a)gmail.com  Reply  Forward Sivan Kinreich, Ph.D. Department of Psychiatry Prediction Modeling and Biomarker Discovery of Brain Disorders lab Henri Begleiter Neurodynamics Laboratory SUNY Downstate Health Sciences University 450 Clarkson Ave. Brooklyn, NY 11203 Tel: 718-270-3322 646-775-8509 Email: sivan.kinreich(a)downstate.edu sivankinreich(a)gmail.com ________________________________ The contents of this email message and any attachments are confidential and are intended solely for the addressee. The information may also be legally privileged. This transmission is sent in trust, for the sole purpose of delivery to the intended recipient. If you have received this transmission in error, any use, reproduction or dissemination of this transmission is strictly prohibited. If you are not the intended recipient, please immediately notify sender by reply email or by (718) 270-HELP.

1 0

Last week of LLM Understanding Summer School.
by Stephen José Hanson June 12, 2024

June 12, 2024

There is incorrect information on the University web site for the LLM understanding, implying that you need a ticket to attend. THIS IS NOT TRUE. the Zoom access is Free; Here's the meeting number; again: ZOOM ACCESS IS FREE<https://uqam.zoom.us/j/85718093402>. MEETING NUMBER IS 85718093402 -- Stephen José Hanson Professor of Psychology Director of RUBIC Member of Exc Comm RUCCS

1 0

13 MSCA-funded PhD positions in (computational) neuroscience and mental health at UAB/CRM (Barcelona)
by Alexandre Hyafil June 12, 2024

June 12, 2024

13 PhD positions are offered by MSCA-COFUND TOUCH <https://webs.uab.cat/touch/> (Towards the next generation of excellent young doctoral researchers on mental health by developing an intersectoral & transdisciplinary approach) at the UAB campus (Barcelona), including two projects at the Computational Neuroscience Group at the CRM: - Modelling circular inference, from the general population to psychotic experience <https://webs.uab.cat/touch/jobs/modelling-circular-inference-from-the-gener…> - Computational modeling of working memory decline in normal aging <https://webs.uab.cat/touch/jobs/computational-modeling-of-working-memory-de…> For informal inquiries about these projects, please contact Klaus Wimmer ( kwimmer(a)crm.cat) or Alex Hyafil (alexandre.hyafil(a)gmail.com) The salary is highly competitive and for 3 years. Applications are due on *June 30* (see guidelines <https://webs.uab.cat/touch/call/>). Eligible candidates must have resided in Spain for less than 12 months in the last 3 years.

1 0

Two weeks left to apply for Travel Grants for the Bernstein Conference 2024
by Alexander Lammers June 12, 2024

June 12, 2024

To support the participation of early career scientists at the Bernstein Conference, the Bernstein Network offers up to 10 Travel Grants, each endowed with 500 €. Details and application: https://bernstein-network.de/bernstein-conference/early-career-scientists/t… ____ BERNSTEIN CONFERENCE Each year the Bernstein Network invites the international computational neuroscience community to the annual Bernstein Conference for intensive scientific exchange. It has established itself as one of the most renowned conferences worldwide in this field, attracting students, postdocs, and PIs from around the world to meet and discuss new scientific discoveries. Conference website: https://bernstein-network.de/bernstein-conference/ ____ IMPORTANT DATES * Travel Grant application deadline: Jun 26, 2024 (15:00 CEST) * Poster abstract submission deadline: Jul 10, 2024 (15:00 CEST) * Early registration deadline: Jul 31, 2024 (23:59 CEST) * Late registration deadline: Sep 11, 2024 (23:59 CEST) * Bernstein Conference: Sep 29 – Oct 2, 2024 ____ INVITED SPEAKERS Dmitriy Aronov (Columbia University, USA) Helen Barron (University of Oxford, UK) Elizabeth Buffalo (University of Washington, USA) Alex Cayco Gajic (Ecole Normale Supérieure, France) Jan Drugowitsch (Harvard University, USA) Jakob Macke (University of Tübingen, Germany) Tirin Moore (Stanford University, USA) Mala Murthy (Princeton University, USA) Memming Park (Champalimaud Foundation, Portugal) Susanne Schreiber (Humboldt Universität zu Berlin, Germany) Xiao-Jing Wang (New York University, USA) ____ CONFERENCE COMMITTEE Matthias Kaschube (Conference Chair) Srdjan Ostojic (Program Chair) Tatiana Engel (Program Vice Chair) Maria Eckstein (Workshop Chair) Wiktor Młynarski (Workshop Vice Chair) & Athena Akrami, Mehrdad Jazayeri, Jennifer Li, Scott Linderman, Ashok Litwin-Kumar, Ida Momennejad, Christina Savin, Andrew Saxe, Daniela Vallentin, Friedemann Zenke. ____ For any further questions, please contact: bernstein.conference(a)fz-juelich.de<mailto:bernstein.conference@fz-juelich.de> ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Jülich GmbH 52425 Jülich Sitz der Gesellschaft: Jülich Eingetragen im Handelsregister des Amtsgerichts Düren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Stefan Müller Geschäftsführung: Prof. Dr. Astrid Lambrecht (Vorsitzende), Karsten Beneke (stellv. Vorsitzender), Prof. Dr. Ir. Pieter Jansens ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------

1 0

INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN HEALTHCARE (AIiH)
by Jiaxiang ZHANG June 11, 2024

June 11, 2024

[apologies for cross-posting] *INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN HEALTHCARE (AIiH)* 4 - 6 September 2024, Swansea, UK https://aiih.cc (*2nd CALL FOR ABSTRACT*) Submission deadline: Monday 1 July (FIRM) *1. INTRODUCTION * https://aiih.cc/abstracts/ AIiH 2024 welcomes late-breaking short abstract submissions to be included as poster or short spot-light presentation at the conference. This is to both enable fast dissemination of promising preliminary findings and encourage attendance by a broader audience of early career researchers (including research students), healthcare professionals, and industrial practitioners. AIiH 2024 is a single-track conference with oral and poster presentations and will include a conference tutorial on generative DNNs and 5 keynote presentations. The conference will also feature a plenary session, which consists of 3 invited speakers who have extensive experience in related areas. AIiH aims to provide a prominent platform for researchers and practitioners who are devoted to improving healthcare using modern artificial intelligence. We recognise that healthcare applications present complex and sometimes unique challenges across a wide spectrum, from ethics to technical developments, that generic AI methods are often inadequate. By creating this dedicated forum, we encourage discussions and disseminations of efficient and effective AI solutions and technologies for healthcare, and in turn we hope to influence the research, technology adoption, and decision making in healthcare. *2. CONFERENCE TUTORIAL* Title: Deep Generative Neural Networks https://aiih.cc/tutorial-generative-dnn/ Dr. Amir Atapour-Abarghouei University of Durham *3. KEYNOTE SPEAKERS* https://aiih.cc/keynote-speakers/ Prof. Hao Ni University College London Dr. Timothy Rittman Cambridge University Dr. Konstantinos Kamnitsas Oxford University Prof. Jacques Fleuriot Edinburgh University Prof. Eiichiro Tanaka Waseda University, Japan *4. PLENARY SESSION ON TRANSLATING AI RESEARCH INTO PRACTICE* https://aiih.cc/plenary-session/ Dr. Alba Di Pardo IRCCS Istituto Neurologico Mediterraneo Neuromed Dr. Noura Al Moubayed Durham University & Evergreen Life Dr. Haoda Fu Eli Lilly and Company *5. ABSTRACT SUBMISSION* Submission instructions are available at the conference website: https://aiih.cc/abstracts/ IMPORTANT DATES Abstract deadline: Monday 1 July (FIRM) Abstract notification: Monday 8 July (FIRM) Abstract registration: Monday 15 July Late registration: Monday 12 August Main conference: Wednesday 4 – Friday 6 September 2024 *6. ORGANISERS* CONFERENCE GENERAL CHAIRS Xianghua Xie, Swansea University, UK Iain Styles, Queen’s University Belfast, UK PROGRAMME CHAIRS Gibin Powathil, Swansea University, UK Marco Ceccarelli, University of Rome Tor Vergata, Italy SPECIAL SESSION CHAIRS Jiaxiang Zhang, Swansea University, UK Rex Ying, Yale University, USA SPONSORSHIP CHAIRS Daniele Cafolla, Swansea University, UK Raoul van Loon, Swansea University, UK PUBLICITY CHAIRS Lu Zhang, Swansea University, UK Beiyu Lin, University of Nevada, USA PUBLICATION CHAIR Jingjing Deng, Durham University, UK *7. SPONSORS* Gold: - Technis Blu - Infordata Silver: - Amicus Therapeutics Bronze: - Zienkiewicz Institute for Modelling, Data and AI - IET Institution of Engineering and Technology Your Researcher Sponsor: - Elsevier Journal of Theoretical Biology Best Paper Award Sponsor: - Springer *8. CONTACT* contact(a)aiih.cc <contact(a)aiih.cc>

1 0