- Comp-neuro - lists.cnsorg.org

LASCON 2016 - VI Latin American School on Computational Neuroscience (Call for applications)
by Antonio C. Roque May 14, 2015

May 14, 2015

**** Apologies for cross-posting **** VI Latin American School on Computational Neuroscience - LASCON 2016 January 3 - 29, 2016, Sao Paulo, Brazil http://sisne.org/lascon APPLICATION DEADLINE: September 6 2015 FIRST ANNOUNCEMENT Dear Colleagues, Following the success of the previous editions of LASCON (Latin American School on Computational Neuroscience), I am pleased to announce the 6th LASCON, which will take place between January 3 - 29 2016 in Sao Paulo, Brazil. LASCON aims at introducing young researchers and graduate and advanced undergraduate students to the use of computational and mathematical methods for modeling neurons and neural networks. These models will be taught via a combination of theoretical lectures and hands-on tutorials with the use of the programs like neuroConstruct, NEURON, NEST, Brian, the Virtual Brain and Matlab. Students will have to work on small research projects (to be done in groups of two), which they will present orally at the end of the school. The following lecturers will be in charge of theoretical classes and tutorials: Arnd Roth, University College, London, UK Benjamin Lindner, Humboldt-Universitaet zu Berlin, Berlin, Germany Boris Marin, University College, London, UK Christophe Pouzat, Universite Paris Descartes, Paris, France Demian Battaglia, Aix-Marseille Universite, Marseille, France German Mato, Centro Atomico Bariloche, San Carlos de Bariloche, RN, Argentina Hannah Bos, Forschungszentrum Juelich, Juelich, Germany Jannis Schuecker, Forschungszentrum Juelich, Juelich, Germany John Murray, New York University, New York, NY, USA Marco Huertas, University of Texas Medical School, Houston, TX, USA Sam Neymotin, State University of New York, New York, NY, USA Volker Steuber, University of Hertfordshire, Hatfield, UK William Lytton, State University of New York, New York, NY, USA There will also be some invited lectures. The current list of invited lecturers is: Andre Kohn, Universidade de Sao Paulo, Sao Paulo, SP, Brazil Antonio Galves, Universidade de Sao Paulo, Sao Paulo, SP, Brazil Cesar Renno-Costa, Universidade Federal do Rio Grande do Norte, Natal, RN, Brazil Hilda Cerdeira, Universidade Estadual Paulista, Sao Paulo, SP, Brazil Julian Tejada, Universidade Federal de Sergipe, Aracaju, SE, Brazil Leonardo Maia, Universidade de Sao Paulo, Sao Carlos, SC, Brazil Marco Idiart, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil Marcelo Reyes, Universidade Federal do ABC, Santo Andre, SP, Brazil Mauro Copelli, Universidade Federal de Pernambuco, Recife, PE, The number of students is limited to 20 and applications should be made electronically via the application form in the school's web page ( http://sisne.org/?page_id=253&lang=en) Applicants are also requested to submit a detailed CV (in English) and to provide two letters of recommendation. The application deadline is September 6 2015 (until midnight, US Pacific time). There is no registration fee for the school. Costs for accommodation will be covered by the school organization. Travel costs have to be covered by the students. Sao Paulo is the major transportation hub in South America, so it is possible to fly from basically anywhere in the world to it. In the selection procedure, priority will be given to Latin-American students, but students from other parts of the world are encouraged to apply as well. LASCON 2016 is an activity of the FAPESP Center for Neuromathematics (NeuroMat -http://neuromat.numec.prp.usp.br) and is jointly organized by the University of Sao Paulo and the Federal University of ABC. I am looking forward to see you in Sao Paulo. Best regards, A. Roque LASCON Director -- Dr. Antonio C. Roque Professor Associado Departamento de Fisica FFCLRP, Universidade de Sao Paulo 14040-901 Ribeirao Preto-SP Brazil - Brasil E-mails: antonior(a)ffclrp.usp.br aroquesilva(a)gmail.com URL: www.sisne.org Tels: +55 16 3315-3768 (sala/office); +55 16 3315-3859 (lab) FAX: +55 16 3315-4887

1 0

Scientist I - Network Analysis Position
by Stefan Mihalas May 13, 2015

May 13, 2015

Dear Colleague, The Allen Institute for Brain Science is embarking on a multiyear project to study neural computation and coding in the mouse cortex. We presently have an opening for a scientist to study structural connectivity of the mouse brain as well as its relation to cell types and mesoscopic measurements of neural activity. For more information or to apply please visit: http://alleninstitute.hrmdirect.com/employment/job-opening.php?req=251378&&… Sincerely, Stefan Mihalas Assistant Investigator T: 206.548.7192 M: 626.376.2414 [cid:image001.png@01D08D7F.230445F0] www.alleninstitute.org<http://www.alleninstitute.org/> Scientist I - Network Analysis This computational neuroscientist will study the organization of structural connectivity networks of the mouse brain. The primary project will focus on network analysis of structural connectivity data, including Allen Mouse Brain Connectivity Atlas. Additional projects will focus on the relation between the structural connectivity and measurements in different modalities. These may include comparisons with cell type characterizations and mesoscopic measurements of neural activity in behaving mice, with the long term goal of uncovering structure-function relation of cortical circuits. RESPONSIBILITES Perform graph theoretical analyses of mouse connectome data for detailed and specific descriptions of network architecture (e.g. hubs, modules, rich club organization) Perform analyses using mouse connectome data to classify connectivity patterns and characterize anatomical rules of cortico-cortical connectivity in mouse brain. Compare organization of mouse cortex with hierarchies described in other species. Perform comparative analyses, computing statistical differences between connectivity data and network architecture collected using different methods, different ages, or in disease models. Compare the connectivity patterns of similar cell types in different structures. Analyze the relation between structural connectivity and other mesocopic brain-wide measurements. Publish/present findings in peer-reviewed journals/scientific conferences. Present a periodic summary of progress to the research community within the organization. QUALIFICATIONS Ph.D. in applied mathematics, computational neuroscience or a related field. Excellent organizational skills and attention to detail. Ability to meet aggressive timelines in a collaborative environment. Strong verbal and written communication skills. Ability to work independently and as a part of a team. Strong publication record.

1 0

Post-doc at Edinburgh
by Matthias Hennig May 13, 2015

May 13, 2015

Dear all We have a vacancy for post-doctoral research on machine learning models applied to neural data. The post is suited to applicants with a machine learning/computational statistics background and an interest in biological applications of statistical modelling. The project is an EPSRC award to develop statistical modelling methodologies for high density neural multi-electrode arrays (HD-MEAs). We will use recordings from a 4,096 channel array to get new insights on the interactions in large populations of neurons. It is led by Guido Sanguinetti and Matthias Hennig. Please get in touch with us for further information and informal inquiries; closing date is 08 June 2015. Details: https://www.vacancies.ed.ac.uk/pls/corehrrecruit/erq_jobspec_version_4.jobs… Best, Matthias -- Matthias H Hennig http://homepages.inf.ed.ac.uk/mhennig/ The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

Multiple open engineering positions in the open-source 3D printed Poppy Robotics project applied to education, at Inria Flowers Lab (Bordeaux, France)
by Pierre-Yves Oudeyer May 13, 2015

May 13, 2015

[Call for application] Multiple open engineering positions in the open-source 3D printed Poppy Robotics project applied to education, at Inria Flowers Lab (Bordeaux, France) Poppy is an open-source platform for the creation, use and sharing of interactive 3D printed robots. It addresses both beginners and experts in the fields of education, science, art, more generally makers. The Poppy platform is based on robust, flexible, easy-to-use and reproduce hardware and software. In particular, the use of 3D printing and rapid prototyping technologies is a central aspect of this project, and makes it easy and fast not only to reproduce the platform, but also to explore morphological variants. The platform, designed at Flowers robotics Lab at Inria and Ensta (France), allows creation and modification of robots of various morphologies, in particular humanoid robot morphologies such as for the robots Poppy Humanoid and Poppy Torso. The platform is also used to create simpler robots with variable morphologies, such as the Poppy Ergo robots, which aim to be at the same time reconfigurable, and easy to construct, plug and play. Web: http://www.poppy-project.org Forum: https://forum.poppy-project.org/categories (600 registered members) Inria Flowers Lab, Bordeaux, France: https://flowers.inria.fr Inria Flowers lab at Inria is now searching for excellent and highly motivated engineers to join the project and develop it towards educational applications, in the context of the project Poppy Education jointly funded by Inria, Région Aquitaine, and European Feder funds. The project Poppy Education targets the application of the Poppy platform to the world of education, both within schools/universities and outside school in networks of FabLabs. As detailed below, we have open positions in software development, web development, as well as design and experimentation of educational content/pedagogical kits, with Scratch-like visual programming tools. Poppy will constitute a major innovative pedagogical platform allowing to cultivate the understanding and creativity of students (and the wider public) to catch up with the current revolution of 3D printing and its integration with computer science, electronics and integrated design methodologies that are now thriving in industry. Within Poppy Education, both the platform, pedagogical content and its use cases will be developed in constant interaction with its end users within schools, universities, FabLabs and other structures acting to educate the wide public to science and technology, within an open-source methodology. Several open positions, described below, are available and will be located within Inria Flowers Lab in Bordeaux, France. Inria Flowers Lab is internationally reknown for its contribution to developmental robotics, on topics such as curiosity-driven learning in robots, imitation learning, robot acquisition of language, cognitive modelling, and applications to education and assistance. Inria: Public science and technology institution established in 1967, Inria is the only public research body fully dedicated to computational sciences. Combining computer sciences with mathematics, Inria’s 3,449 researchers strive to invent the digital technologies of the future. Educated at leading international universities, they creatively integrate basic research with applied research and dedicate themselves to solving real problems, collaborating with the main players in public and private research in France and abroad and transferring the fruits of their work to innovative companies. The researchers at Inria published over 4,500 articles in 2013. They are behind over 270 active patents and 110 start-ups. In 2013, Inria’s budget came to 235 million euros, 25% of which represented its own resources. Flowers Lab is located at the Inria Bordeaux site: http://www.inria.fr/en/centre/bordeaux/overview/discover-our-centre How to apply: send an email with your CV AND letter of motivation to poppy-jobs(a)inria.fr indicating the CODE of the mission for which you apply (see codes below). Detailed information and documents on projects realized by candidates will be appreciated. Code: Poppy Education 1 Job title: Engineer in software development for integrated development platform Experience: up to 8 years after engineering diploma (plutôt 2 à 8 ans d'experience ? je suis pas sûr qu'on veuille un IJD) Duration: 12 to 24 months Mission Develop and integrate software programming tools that allow users in education to discover and learn programming and robotics Job description: Extend and maintain the pypot python programming library (https://github.com/poppy-project/pypot) Extend and maintain the integration of pypot with Ipython and Jupyter Extend and maintain the integration of programming tools with physical Poppy robots and a simulator Develop installation tools allowing rapid "one-click" installation of programming tools on several platforms including Raspberry Pi, Windows, Mac Os, Linux Document developped tools Participate to animating the users community, both online through the forum and web site and through participating to live educational robotics events Candidate profile: Extended python programming experience Experience in embedded programming (e.g. Rasberry Pi, Arduino, Embedded Linux) Rich programming skills including javascript, C++ Multiplatform software installers technologies (Windows/MacOS/Linux) Experience of programming robots User-centered methodology for code development Experience in open-source projects Strong experience with versionning tool (Git) Excellent programming practices Code: Poppy Education 2 Job title: Engineer in web software technologies Experience: Up to 2 years after engineering diploma Duration: 12 to 24 months Mission Integration and development of web software technologies for the Poppy web sites Job description: Develop Wordpress plugins for http://www.poppy-project.org Develop plugins and personalization of Discourse-based Poppy forum: https://forum.poppy-project.org Integrate a wiki technology for Poppy Documentation Unify graphical design of web sites Setup maintainance methods for Poppy web sites (including update of supporting softwares) Document developped tools Daily update of web site content and animation of the community in the forum Candidate profile: Excellent javascript skills Experience in developping plugins for Wordpress Experience in web design and web site development Experience in open-source projects Strong experience with versionning tool (Git) Excellent programming practices Code: Poppy Education 3 Job title: Engineer in software development for web-based applications Experience: Up to 2 years after engineering diploma (+0 to +2 experience) Duration: 12 to 24 months Mission Develop web applications for educational robotics Job description: Develop web applications allowing the configuration of robots and checking the state of motors, sensors and electronics (accessible from browsers on computers and tablets) Extend and maintain a REST API for the pypot python programming library Develop a web applications allowing basic control of Poppy robots (launching behaviours, teleguiding robots) Extend and develop visual programming tools based on Snap! language (including developping extensions to Snap!) Develop compatibility with Scratch programming language Develop an app store allowing easy community sharing of programs written in python or visual programming languages Participate to animating the users community, both online through the forum and web site and through participating to live educational robotics events Candidate profile: Excellent programming skills in javascript and python Experience in web app architecture design Experience in interface design and user-centered methodology Experience in web communication technologies and APIs Experience in responsive and mobile friendly web design Strong experience with versionning tool (Git) Excellent programming practices Code: Poppy Education 4 Job title: Engineer for pedagogical design and experimentation of educational activities in robotics Experience: up to 8 years after engineering/university diploma (pareil +2, autant pas avoir un IJD) Duration: 12 to 24 months Mission Design, experiment and disseminate educational activities based on the Poppy robotic platform Job description: Analyze the educational needs of several populations of users (high-school students, university students, FafLab members, ...) Design pedagogical sequences and scenarios of activities targeted to different populations of students, based on visual programming and python programming tools Initiate contacts with users for experimentation Evaluate pedagogical activities in real contexts of use (in schools, universities and FabLabs) Produce integrated polished multimedia documents embedding "ready to use" pedagogical content Produce multimedia content to publicize educational uses of the platform Participate to events to promote educational uses of the platform Candidate profile: Experience in teaching computer science and technology Experience in user-centered methodologies Experience in python programming language (or other scripting languages) Familiarity with Scratch like visual, drag-and-drop programming language Interest in interdiciplinary collaborations (education, science, art, makers) Creative thinking to design motivating learning scenarios Code: Poppy Education 5 Job title: Engineer for coordination of pedagocial activities and events in Poppy Project Experience: senior level pedagogical expert Mission Coordinate the design and dissemination of pedagogical activitives, organization of educational dissemination events Job description: Build a focus group of pedagogical experts to design the specification of educational robotics activities Build a network of educational partner institutions to welcome experimentation of educational activities Coordinate the organization of experimentation and dissemination of pedagogical activities Design pedagogical sequences and scenarios of activities targeted to different populations of students, based on visual programming and python programming tools Candidate profile: Extended experience of teaching with/about digital technologies/ICT Extended experience of collaboration with educational institutions Experience in designing pedagocical content for various age ranges Knowledge of French education programs at high-school levels Pierre-Yves Oudeyer Research director, Inria Head of Flowers Laboratory Inria and Ensta Paris-Tech http://www.pyoudeyer.com https://flowers.inria.fr

1 0

International Brain Connectivity Workshop, June 10-12, 2015 - last call
by Scott Makeig May 13, 2015

May 13, 2015

*14th International Workshop on Brain Connectivity* The international Workshop on Brain Connectivity (WBC), to be held at the University of California San Diego (June 10-12, 2015), will focus for the first time primarily on the role of electrophysiological data in understanding the fine temporal scale underlying brain network dynamics as well as the mechanisms by which emergent dynamics originate from multi-scale interactions between brain areas, brain activity frequencies and timescales, and between brain activity and behavior. This focus reflects the growing realization that to understand how the brain supports behavior and experience requires imaging of distributed brain activity with fine temporal resolution. The first day of the Workshop will be dedicated to lectures, followed by two days of half-day workshop sessions. The WBC will bring together leading investigators in the fields of neuroimaging, cognitive and computational neuroscience to discuss current and forthcoming challenges for brain connectivity research based on electrophysiological data. Speakers include Klaas Stephen (Zurich), Carsten Wolters (Muenster), Eugene Izhekevich (San Diego), Stephan Kiebel (Leipzig), Matias Palva (Helsinki), Alain Destexhe (France), Charles Schroeder (Columbia U), Ryan Canolty (U Houston), Stefan Haufe (New York), M. Muthuram (Germany), Kathleen Poston (Stanford) and others. Its goals are to stimulate research students and scientists from diverse backgrounds for research in topics related to electrophysiology and brain connectivity, to promote larger international collaborations in these fields, and to inform the neuroscience community about emerging methods, and comprehensive theory and models for brain connectivity analysis, in particular from electrophysiological recordings. - for the organizers, Scott Makeig, Swartz Center, UCSD Registration (USD) Postdocs, Academics, Others 400 Students (Graduate and Undergraduate) 260 For more information and registration, visit *http://sccn.ucsd.edu/BrainConnectivityWorkshop2015 <http://sccn.ucsd.edu/BrainConnectivityWorkshop2015>* -- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott -- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott -- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott -- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott

1 0

[journals] CFP: IEEE ToH Issue on Active Touch Sensing
by yon May 13, 2015

May 13, 2015

A reminder of the June 1 deadline for the forthcoming special issue of the IEEE Transactions on Haptics on “Active Touch Sensing in Robots, Humans, and Other Animals". We hope you will consider contributing. Please share this announcement with your colleagues. Thank you! Yon Visell Call for Papers IEEE Transactions on Haptics Special Issue on Active Touch Sensing in Robots, Humans, and Other Animals Submission deadline (EXTENDED): June 1 2015 First decision to authors: August 1 2015 Complete Details: http://re-touch-lab.com/ieee-toh-issue <http://re-touch-lab.com/ieee-toh-issue> Guest Editors: Yon Visell, Drexel University Vincent Hayward, Université Pierre et Marie Curie Mitra Hartmann, Northwestern University Nathan Lepora, University of Bristol Topical Summary: This special issue addresses the challenges posed by actively sensing and interacting with the world through the sense of touch, whether the latter is implemented through a technological or biological system. Active touch sensing is recovering information about the world by ‘touching’ rather than ‘being touched’ – by interpreting signals from sensors whose motion is deliberately controlled to facilitate information gain. The scope of this issue includes both biological and technological systems for active touch sensing, and implications for haptics. This issue will consider electronic systems for active touch sensing that are biologically inspired systems, in addition to other inherently active approaches to touch sensing. Biological systems for active touch sensing are highly capable, and, by comparison, the field of robotic touch sensing is in its infancy. The former demonstrate many valuable concepts for active touch sensing that are being intensively investigated. They have also illustrated ways that active touch sensing is enabled through specialized sensory transduction channels, biomechanics, structural morphology, behavioral, and control strategies that are implemented by biological systems, and through other advantages that they achieve, including robustness, adaptability, and power efficiency. Similar challenges must be overcome if technological systems are to one day achieve comparable levels of sensorimotor performance to biological systems. Specific topics of interest include: - The design of electronic or biological systems for active touch sensing, their principles of operation (electronic, biological, mechanical) and performance. - Biomechanical aspects of active touch sensing. - Relations between active touch sensing and behavior, especially motor activity. - Sensory information processing and sensorimotor control strategies in humans, animals, or robots related to active touch sensing, including behavioral, physiological or neuroscience investigations, and neural control. - Learning by active touch. - Applications to the design of haptic interfaces. Authors should specifically include a discussion of how their results shed light on fundamental aspects of active touch sensing rather than assuming that this can be inferred from context. Submission process: Visit http://www.computer.org/toh <http://www.computer.org/toh> to view formatting requirements, and submit your paper at https://mc.manuscriptcentral.com/th-cs <https://mc.manuscriptcentral.com/th-cs>. When uploading your paper please select the appropriate special issue title under the category “Manuscript Type.” Further detail: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6901353 <http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6901353> _________________________________________________________ Yon Visell, PhD Assistant Professor Drexel University | Electrical and Computer Engineering Philadelphia, USA www.re-touch-lab.com <http://www.re-touch-lab.com/> Office: +1 215 571 4507 Mobile: +1 267 800 8960 _________________________________________________________

1 0

Postdoctoral position - in vivo multiphoton imaging
by Simon Schultz May 13, 2015

May 13, 2015

Job Title Research Associate –Two Photon Imaging of Alzheimer’s Disease Department/Division/Faculty Department of Bioengineering - Centre for Neurotechnology Salary Range: £33,410 - £42,380 per annum South Kensington Campus Full time, fixed-term for 36 months The Centre for Neurotechnology forms part of the cross-departmental Institute of Biomedical Engineering at Imperial College London. The Centre, in association with the Department of Bioengineering, hosts world-class experimental facilities for research at the interface between neuroscience and engineering. The Centre has recently initiated a philanthropically funded project on the application of biomedical engineering principles to understand how the progression of Alzheimer’s Disease affects information processing in cortical circuits, and test new therapeutic interventions. Applications are invited for the above post to work with Dr Simon Schultz on in vivo two photon imaging of calcium signals in the dendrites of neocortical pyramidal cells. The aim of this work is to understand how synaptic integration changes during the progression of Alzheimer’s Disease, and to test the effects on dendrites of new therapeutic strategies being developed for the treatment of Alzheimer’s. Candidates must be in possession of a doctoral degree or equivalent in neuroscience, physics, engineering, or similar. Prior experience with in vivo two photon imaging is a requirement. Preference will be given to candidates with both a strong experimental background, and experience in quantitative neurophysiological data analysis using MATLAB. This post will be based at the South Kensington Campus. Closing Date 14 June 2015 (midnight BST) Email s.schultz(a)imperial.ac.uk for any queries about the position. How To Apply Our preferred method of application is online via http://www.imperial.ac.uk/job-applicants/ Please submit a CV as well as an application form for this post.

1 0

Comp Neuro List Submission for jobs
by Ben Hayden May 13, 2015

May 13, 2015

The Hayden Lab at the University of Rochester is looking to hire two post-docs in systems and/or computational neuroscience. Our lab investigates the neuronal mechanisms of economic choice, reward cognition, and self-control. We are especially interested in the cingulate cortex, the orbital surface, and the striatum. We use simple gambling tasks, naturalistic foraging tasks and game theoretic decisions, including multi-monkey interaction. We are particularly interested in devising and testing computational models of economic choice, foraging decisions, and self-control, using attractor network, PDP network, and swarm cognition models. We focus on non-human primate recordings, although we are also looking for a purely theoretical post-doc as well. Prior experience in computer programming or statistics highly recommended. The successful candidate will build a portfolio either purely in theoretical/computational neuroscience or in single unit physiology. The University of Rochester is an equal opportunity employer. Standard NIH salary range applies. Lab website: http://haydenlab.com If interested, send an email to benhaydenlab(a)gmail.com.

1 0

Post-doc fellowship on computational neurosciences and stereoscopic vision
by cottereau＠cerco.ups-tlse.fr May 12, 2015

May 12, 2015

Post-doc fellowship with Benoit Cottereau on computational neurosciences and stereoscopic vision Centre de Recherche Cerveau et Cognition, Toulouse, France Starting date: September December 2015 A postdoctoral research position is available to work with Benoit Cottereau (CNRS Researcher) in the ECO-3D team of the CerCo laboratory (Toulouse, http://www.cerco.ups-tlse.fr/?lang=en) The aim of the project is to develop new models of stereoscopic vision in primate that automatically become selective to the most frequently occurring 3D properties of their visual environment (e.g. 3D motion direction). These models will possibly be validated through experimental data (psychophysics and brain imaging in both human and non-human primates). The applicant should hold a solid background in computational neurosciences, signal processing as well as skills in programming. Knowledge on vision in primate would be highly appreciated. The city of Toulouse is an attractive city with high quality of life located in the south west of France (close to the Pyrenees, the Mediterranean sea and the Spanish border). The position is for 18 months with standard French salaries. Applications should be sent to benoit.cottereau(a)cerco.ups-tlse.fr including a CV and 2 names of references.

1 0

5th Bernstein Sparks Workshop: Neural models of decision making in natural inference tasks - from theory to experiment
by Ralf M. Haefner May 12, 2015

May 12, 2015

5th Bernstein Sparks Workshop: Neural models of decision making in natural inference tasks - from theory to experiment June 11-12, 2015 in Tübingen, Germany Abstract submission and registration deadline: May 20, 2015 (attendance limited: first come, first served) Detailed information at http://www.bccn-tuebingen.de/events/5th-bernstein-sparks-workshop-2015.html Confirmed Speakers: • Wieland Brendel (University of Tuebingen) • David Cox (Harvard University) • Jaime de la Rocha (IDIBAPS, Barcelona) • Nicolas Heess (Google DeepMind) • Mate Lengyel (University of Cambridge) • Wolfgang Maass (Technical University Graz) • Hendrikje Nienborg (University of Tuebingen) • Andreas Tolias (BCM Houston) • Dan Yamins (MIT) • Rich Zemel (University of Toronto) The brain needs to extract behaviorally relevant information from sensory signals which contain only indirect, incomplete and highly variable information about the world. Beginning with the study of the neural basis of random dot motion discrimination in the late eighties, neural decision making has been studied extensively in neuroscience. However, the models used to reason about decision making do not account for the computational complexity of natural inference tasks such as object recognition, speech recognition or alike. Based on recent advances in machine learning more and more complex artificial neural networks are developed that become increasingly proficient in mimicking perceptual inference abilities of humans and animals. As a side effect of their popularity in technology, the increasing availability and diversity of high-performing neural network models opens a new door for studying the neural mechanisms of robust decision making. Important differences between these networks are the presence or absence of feedback connections, the presence or absence of stochasticity, and the diversity of different nonlinear mechanisms. The existence of this diversity mirrors important discussions in neuroscience on the role and effect of feedback signals is in the brain [1], whether the brain represents and computes with probabilities [2], whether feedback signals are essential for performing probabilistic inference in hierarchical models [3], and whether neural stochasticity can be interpreted in terms of sampling [4] or regularization such as dropout [5]. As a joint effort between theoreticians and experimentalists, the goal of this Sparks workshop will be to survey and discuss the role of these mechanisms for robust decision making in artificial and real neural networks and to derive discriminative experimental tests and tools that seem most promising to analyze them. [1] Gilbert & Li, Nat Rev Neuro 2013 [2] Pouget et al, Nat Rev Neuro 2013 [3] Lee & Mumford, JOSA 2003 [4] Fiser et al., TICS 2010 [5] Srivastava et al, JMLR 2014 Organizers: • Matthias Bethge (BCCN Tübingen) • Ralf Haefner (University of Rochester) • Richard Hahnloser (University of Zürich) • Bernstein Coordination Site (BCOS)

1 0