BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:US/Eastern BEGIN:DAYLIGHT DTSTART:20180311T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20171105T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20180503T000000Z UID:87D7985F-5A69-4E46-BEF8-E5A3C32C4F79 DTSTART;TZID=US/Eastern:20180205T190000 DTEND;TZID=US/Eastern:20180205T203000 DESCRIPTION:During our february monthly event\, Dr. Matthew Phillips from K itware will present about the computer vision topic involving deep learnin g - understanding objects and scenes with Generative Adversarial Networks( GANs).\n\nGenerative adversarial networks (GANs) are a class of artificial intelligence algorithms used in unsupervised machine learning\, implement ed by a system of two neural networks contesting with each other in a zero -sum game framework.\n\nIn simple words\, GANs are artificial neural netwo rks that work together to give better answers. One neural network is the t ricky network\, and the other one is the useful network. The tricky networ k will try to give an input to the useful network that will cause the usef ul network to give a bad answer. The useful network will then learn not to give a bad answer\, and the tricky network will try to trick the useful n etwork again. As this continues\, the useful network will get better and n ot become tricked as often\, and the useful network will be able to be use d to make good predictions.\n\nGANs provide a way to learn deep representa tions without extensively annotated training data. They achieve this throu gh deriving backpropagation signals through a competitive process involvin g a pair of networks.\n\nGANs have received wide attention in the machine learning field because of their potential to learn high-dimensional\, comp lex real data. Specifically\, they do not perform distribution assumptions and can simply infer real-like samples from latent space. This powerful p roperty leads GANs to be applied to various applications such as image syn thesis\, image attribute editing\, image translation\, image super-resolut ion and classification.\n\nBio of the speaker:\n\nDr. Matthew Phillips rec eived a B.A. in philosophy and mathematics from Tufts University\, and he received a Ph.D. in philosophy with a certificate in cognitive science fro m Rutgers University. After he completed his Ph.D.\, Dr. Phillips moved in to neuroscience\, where he first focused on visual/oculomotor psychophysic s. He later shifted to primate cellular electrophysiology and\, finally\, to computational neuroscience. His focus on computational neuroscience bro ught him to Duke University and the Research Triangle Park (RTP). In the R TP\, Matt worked briefly as a C++ engineer and pursued side projects in ma chine learning. He now works full time in machine learning and computer vi sion at Kitware.\n\nDr. Phillips has received numerous awards\, grants and fellowships. In 2005\, he received the James S. McDonnell Foundation 21st Century Postdoctoral Fellowship Award\, which he declined. A year later\, he received a Fight for Sight fellowship. Then\, in 2008\, he accepted an award to present at the Advances in Computational Motor Control meeting. In the same year\, he began a Postdoctoral Individual National Research Se rvice Award from the National Eye Institute of the National Institutes of Health for “Using saccadic adaptation to probe the coordinate system of parietal neurons.” Later\, in 2012\, he received a research grant from t he American Academy of Neurology for “Assessing efficiency of learning t he neurologic exam with a visual tracking device” with co-principal inve stigator James Noble\, MD.\n\nDr. Phillips has also contributed to conside rable applications development and open source projects including signal p rocessing algorithms\, neural waveforms and computer vision based analytic s.\n\nPlease refer to the following link to know more about Dr. Matthew Ph illips\,\n\nhttps://www.kitware.com/matthew-phillips/\n\nIEEE ENCS RA24 ch apter appreciates the passion\, drive\, highly impressive efforts from Dr. Matthew Phillips and wishes the very best to his career & life.\n\nAgenda : \n6:20-7:00pm Networking with pizza and soda\n\nThe times below are appr oximate and are given just as a guideline:\n\n7:00-7:10pm News and announc ements\n\n7:10-7:50pm Dr. Matthew Phillips session about techniques unders tanding objects and scenes with Generative Adversial Networks.\n\n7:50-8:0 0pm Brief Q/A from IEEE students\, hobbyists\, professionals and seniors\n \n8:00-8:30pm Show-n-Tell (if interested\, bring out your projects to show to other members)\n\nRoom: 1007\, Bldg: Engineering Building 1 (EB1)\, 91 1 Partners Way\, Raleigh\, North Carolina\, United States\, 27606 LOCATION:Room: 1007\, Bldg: Engineering Building 1 (EB1)\, 911 Partners Way \, Raleigh\, North Carolina\, United States\, 27606 ORGANIZER:mbalasu@ncsu.edu SEQUENCE:4 SUMMARY:Tripping the light GANtastic: Understanding objects and scenes with Generative Adversarial Networks - Dr. Matthew Phillips URL;VALUE=URI:https://events.vtools.ieee.org/m/160206 X-ALT-DESC:Description: <br /><p>During our february monthly event\, <stron g>Dr. Matthew Phillips </strong>from Kitware will present about the comput er vision topic involving deep learning -<strong> understanding objects an d scenes with </strong><strong>Generative Adversarial Networks(GANs). </st rong></p>\n<p><strong>Generative adversarial networks (GANs)</strong>&nbsp \;are a class of artificial intelligence algorithms used in unsupervised m achine learning\, implemented by a system of two neural networks contestin g with each other in a zero-sum game framework.</p>\n<p>In simple words\,< strong> GANs</strong>&nbsp\;are artificial neural networks that work toget her to give better answers. One neural network is the tricky network\, and the other one is the useful network. The tricky network will try to give an input to the useful network that will cause the useful network to give a bad answer. The useful network will then learn not to give a bad answer\ , and the tricky network will try to trick the useful network again. As th is continues\, the useful network will get better and not become tricked a s often\, and the useful network will be able to be used to make good pred ictions.</p>\n<p><strong>GANs</strong> provide a way to learn deep represe ntations without extensively annotated training data. They achieve this th rough deriving backpropagation signals through a competitive process invol ving a pair of networks.</p>\n<p><strong>GANs</strong> have received wide attention in the machine learning field because of their potential to lear n high-dimensional\, complex real data. Specifically\, they do not perform distribution assumptions and can simply infer real-like samples from late nt space. This powerful property leads GANs to be applied to various appli cations such as image synthesis\, image attribute editing\, image translat ion\, image super-resolution and classification.</p>\n<p><strong><u>Bio of the speaker:</u></strong></p>\n<p><strong>Dr. Matthew Phillips </strong>r eceived a B.A. in philosophy and mathematics from Tufts University\, and h e received a Ph.D. in philosophy with a certificate in cognitive science f rom Rutgers University. <strong>After he completed his Ph.D.\, Dr. Phillip s moved into neuroscience\, where he first focused on visual/oculomotor ps ychophysics. He later shifted to primate cellular electrophysiology and\, finally\, to computational neuroscience. His focus on computational neuros cience brought him to Duke University and the Research Triangle Park (RTP) . In the RTP\, Matt worked briefly as a C++ engineer and pursued side proj ects in machine learning. He now works full time in machine learning&nbsp\ ;and computer vision&nbsp\;at Kitware.</strong></p>\n<p><strong>Dr. Philli ps has received numerous awards\, grants and fellowships. In 2005\, he rec eived the James S. McDonnell Foundation 21st Century Postdoctoral Fellowsh ip Award\, which he declined. A year later\, he received a Fight for Sight fellowship. Then\, in 2008\, he accepted an award to present at the Advan ces in Computational Motor Control meeting. In the same year\, he began a Postdoctoral Individual National Research Service Award from the National Eye Institute of the National Institutes of Health for &ldquo\;Using sacca dic adaptation to probe the coordinate system of parietal neurons.&rdquo\; Later\, in 2012\, he received a research grant from the American Academy of Neurology for &ldquo\;Assessing efficiency of learning the neurologic e xam with a visual tracking device&rdquo\; with co-principal investigator J ames Noble\, MD.</strong></p>\n<p>Dr. Phillips has also contributed to con siderable applications development and open source projects including sign al processing algorithms\, neural waveforms and computer vision based anal ytics.</p>\n<p>Please refer to the following link to know more about Dr. M atthew Phillips\,</p>\n<p><strong>https://www.kitware.com/matthew-phillips /</strong></p>\n<p>IEEE ENCS RA24 chapter appreciates the passion\, drive\ , highly impressive efforts from Dr. Matthew Phillips and wishes the very best to his career &amp\; life.</p><br /><br />Agenda: <br /><p>6:20-7:00p m Networking with pizza and soda</p>\n<p>The times below are approximate a nd are given just as a guideline:</p>\n<p>7:00-7:10pm News and announcemen ts</p>\n<p>7:10-7:50pm Dr. Matthew Phillips session about techniques under standing objects and scenes with Generative Adversial Networks.</p>\n<p>7: 50-8:00pm Brief Q/A from IEEE students\, hobbyists\, professionals and sen iors</p>\n<p>8:00-8:30pm Show-n-Tell (if interested\, bring out your proje cts to show to other members)</p> END:VEVENT END:VCALENDAR