BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/Madrid BEGIN:DAYLIGHT DTSTART:20170326T030000 TZOFFSETFROM:+0100 TZOFFSETTO:+0200 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3 TZNAME:CEST END:DAYLIGHT BEGIN:STANDARD DTSTART:20171029T020000 TZOFFSETFROM:+0200 TZOFFSETTO:+0100 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:CET END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20170609T110340Z UID:4B81A9BB-4D03-11E7-8752-0050568D2FB3 DTSTART;TZID=Europe/Madrid:20170614T150000 DTEND;TZID=Europe/Madrid:20170614T170000 DESCRIPTION:Next June 14\, Dr. Simon Thorpe and Dr. Timothee Masquelier\, f rom CNRS\, France\, will give two talks about Maching Learning as part of the PhD program on Physical Sciences and Technology of the University of S eville and co-sponsored by the Spain Chapter of the IEEE Circuits and Syst ems Society. The talks will take place at the Institute of Microelectronic s of Seville\, IMSE (CSIC/University of Seville)\, Seville\, Spain\, on We dnesday\, June 14\, from 15:00h to 17:00h.\n\nTitle\, abstract and lecture rs' biographies:\n\nSimon Thorpe: Finding Repeating Structures – the Sec ret of Intelligence?\n\nTimothee Masquelier: Spike-based computing and lea rning in brains\, machines\, and visual systems in particular\n\nAbstract: The two talks cover the field of brain-inspired learning for artificial m achine learning algorithms\, with special emphasis on vision systems. Brai ns compute with spikes in a fashion that information seems to be encoded i n a highly efficient manner for minimum power\, but still with outstanding computing capabilities compared to human made "intelligent" machines. In these talks we will expose how the spiking nature of nervous information e ncoding can be exploited for learning tasks\, with outreach to artificial machine learning systems.\n\nUsing simulations\, we have first shown that\ , thanks to the physiological learning mechanism referred to as spike timi ng-dependent plasticity (STDP)\, neurons can detect and learn repeating sp ike patterns\, in an unsupervised manner\, even when those patterns are em bedded in noise\, and the detection can be optimal. Importantly\, the spik e patterns do not need to repeat exactly: it also works when only a firing probability pattern repeats\, providing this profile has narrow (10-20ms) temporal peaks. Brain oscillations may help in getting the required tempo ral precision\, in particular when dealing with slowly changing stimuli. A ll together\, these studies show that some envisaged problems associated t o spike timing codes\, in particular noise-resistance\, the need for a ref erence time\, or the decoding issue\, might not be as severe as once thoug ht. These generic STDP-based mechanisms are probably at work in particular the visual system\, where they can explain how selectivity to visual prim itives emerges\, leading to efficient object recognition systems. High spi ke time precision is required\, and microsaccades could help.\n\nBiographi es\n\nSimon Thorpe is a CNRS Research Director (recently promoted to Class e Exceptionnelle) who studied Physiology\, Psychology & Philosophy (PPP) a t Oxford (graduating in 1977)\, got his PhD with Edmund Rolls\, did a post doc in Canada (with Max Cynader) and then came to France in 1982. Recruite d by the CNRS in 1983\, he moved from Paris to Toulouse in 1993 to help cr eate the Brain & Cognition Research Center (CerCo). He became the lab dire ctor (taking over from Michèle Fabre-Thorpe) in 2014\, and also took over the direction of the ISCT from François Chollet in January 2016.\n\nHe i s very keen on interdisciplinary research\, and does a mix of neurophysiol gy\, psychophyiscs\, computer modeling and theoretical work. He is current ly half way through a 5 year ERC advanced grant called “M4 – memory me chanisms in man and machine”\, which aims to understand how we can store sensory memories that can last for an entire lifetime. His hypothesis is that we store memories in “grandmother cells” that can remain totally silent for months or years – neocortical dark matter.\n\nTimothee Masque lier is a researcher in computational neuroscience. His research is highly interdisciplinary - at the interface between biology\, computer science\, and physics. He uses numerical simulations and analytical calculations to gain understanding on how the brain works\, and more specifically on how neurons process\, encode and transmit information through action potential s (a.k.a spikes)\, in particular in the visual modality. He is also intere sted in bio-inspired computer vision and neuromorphic engineering. He was trained at Ecole Centrale Paris (Ingénieur 1999)\, MIT (M. Sc. 2001)\, an d Univ. Toulouse 3 (PhD 2008). He was recruited by the CNRS in 2012.\n\nCo -sponsored by: Instituto de Microelectrónica de Sevilla\, IMSE-CNM (CSIC / Universidad de Sevilla)\n\nSpeaker(s): \, \, \, \n\nInstituto de Microel ectrónica de Sevilla\, Parque Tecnológico de la Cartuja\, C/ Américo Ve spucio S/N\, Sevilla\, Andalucia\, Spain LOCATION:Instituto de Microelectrónica de Sevilla\, Parque Tecnológico de la Cartuja\, C/ Américo Vespucio S/N\, Sevilla\, Andalucia\, Spain ORGANIZER:jdelarosa@ieee.org SEQUENCE:0 SUMMARY:Invited Talks on Machine Learning at IMSE-CNM (CSIC /University of Seville) URL;VALUE=URI:https://events.vtools.ieee.org/m/45848 X-ALT-DESC:Description: <br /><p>Next June 14\, Dr. Simon Thorpe and Dr. &n bsp\;Timothee Masquelier\, from CNRS\, France\, will give two talks about Maching Learning as part of the PhD program on Physical Sciences and Techn ology of the University of Seville and co-sponsored by the Spain Chapter o f the IEEE Circuits and Systems Society. The talks will take place at the Institute of Microelectronics of Seville\, IMSE (CSIC/University of Sevill e)\, Seville\, Spain\, on Wednesday\, June 14\, from 15:00h to 17:00h.</p> \n<p><strong>Title\, abstract and lecturers' biographies:</strong></p>\n<p >&nbsp\;</p>\n<p><strong>Simon Thorpe: <em>Finding Repeating Structures &n dash\; the Secret of Intelligence?</em></strong></p>\n<p><strong>Timothee Masquelier: <em>Spike-based computing and learning in brains\, machines\, and visual systems in particular</em></strong></p>\n<p>&nbsp\;</p>\n<p><st rong>Abstract</strong>:&nbsp\;The two talks cover the field of brain-inspi red learning for artificial machine learning algorithms\, with special emp hasis on vision systems. Brains compute with spikes in a fashion that info rmation seems to be encoded in a highly efficient manner for minimum power \, but still with outstanding computing capabilities compared to human mad e "intelligent" machines. In these talks we will expose how the spiking na ture of nervous information encoding can be exploited for learning tasks\, with outreach to artificial machine learning systems.</p>\n<p>Using simul ations\, we have first shown that\, thanks to the physiological learning m echanism referred to as spike timing-dependent plasticity (STDP)\, neurons can detect and learn repeating spike patterns\, in an unsupervised manner \, even when those patterns are embedded in noise\, and the detection can be optimal. Importantly\, the spike patterns do not need to repeat exactly : it also works when only a firing probability pattern repeats\, providing this profile has narrow (10-20ms) temporal peaks. Brain oscillations may help in getting the required temporal precision\, in particular when deali ng with slowly changing stimuli. All together\, these studies show that so me envisaged problems associated to spike timing codes\, in particular noi se-resistance\, the need for a reference time\, or the decoding issue\, mi ght not be as severe as once thought. These generic STDP-based mechanisms are probably at work in particular the visual system\, where they can expl ain how selectivity to visual primitives emerges\, leading to efficient ob ject recognition systems. High spike time precision is required\, and micr osaccades could help.</p>\n<p><strong>Biographies</strong></p>\n<p><strong >Simon Thorpe</strong> is a CNRS Research Director (recently promoted to C lasse Exceptionnelle) who studied Physiology\, Psychology &amp\; Philosoph y (PPP) at Oxford (graduating in 1977)\, got his PhD with Edmund Rolls\, d id a postdoc in Canada (with Max Cynader) and then came to France in 1982. Recruited by the CNRS in 1983\, he moved from Paris to Toulouse in 1993 t o help create the Brain &amp\; Cognition Research Center (CerCo). He becam e the lab director (taking over from Mich&egrave\;le Fabre-Thorpe) in 2014 \, and also took over the direction of the ISCT from Fran&ccedil\;ois Chol let in January 2016.</p>\n<p>He is very keen on interdisciplinary research \, and does a mix of neurophysiolgy\, psychophyiscs\, computer modeling an d theoretical work. He is currently half way through a 5 year ERC advanced grant called &ldquo\;M4 &ndash\; memory mechanisms in man and machine&rdq uo\;\, which aims to understand how we can store sensory memories that can last for an entire lifetime. His hypothesis is that we store memories in &ldquo\;grandmother cells&rdquo\; that can remain totally silent for month s or years &ndash\; neocortical dark matter.</p>\n<p><strong>Timothee Masq uelier</strong> is a researcher in computational neuroscience. His researc h is highly interdisciplinary - at the interface between biology\, compute r science\, and physics. He uses numerical simulations and analytical calc ulations to gain understanding on how the brain works\, and more specifica lly on how neurons process\, encode and transmit information through actio n potentials (a.k.a spikes)\, in particular in the visual modality. He is also interested in bio-inspired computer vision and neuromorphic engineeri ng. He was trained at Ecole Centrale Paris (Ing&eacute\;nieur 1999)\, MIT (M. Sc. 2001)\, and Univ. Toulouse 3 (PhD 2008). He was recruited by the C NRS in 2012.</p> END:VEVENT END:VCALENDAR