BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Toronto BEGIN:DAYLIGHT DTSTART:20210314T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20211107T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20210625T003517Z UID:1D8BBF8C-8277-4366-8A8A-C9D8620B0023 DTSTART;TZID=America/Toronto:20210624T110000 DTEND;TZID=America/Toronto:20210624T123000 DESCRIPTION:Abstract\n\nThe application of supervised learning techniques f or the design of the physical layer of a communication link is often impai red by the limited amount of pilot data available for each device\; while the use of unsupervised learning is typically limited by the need to carry out a large number of training iterations. In this talk\, meta-learning\, or learning-to-learn\, is introduced as a tool to alleviate these problem s. The talk will consider an Internet-of-Things (IoT) scenario in which de vices transmit sporadically using short packets with few pilot symbols ove r a fading channel. The number of pilots is generally insufficient to obta in an accurate estimate of the end-to-end channel\, which includes the eff ects of fading and of the transmission-side distortion. To tackle this pro blem\, pilots from previous IoT transmissions are used as meta-training da ta in order to train a demodulator that is able to quickly adapt to new en d-to-end channel conditions from few pilots. Various state-of-the-art meta -learning schemes are adapted to the problem at hand and evaluated\, inclu ding MAML\, FOMAML\, REPTILE\, and CAVIA. Both offline and online solution s are developed.\n\nBiography\n\nOsvaldo Simeone is a Professor of Informa tion Engineering with the Centre for Telecommunications Research at the De partment of Engineering of King's College London\, where he directs the Ki ng's Communications\, Learning and Information Processing lab. He received an M.Sc. degree (with honors) and a Ph.D. degree in information engineeri ng from Politecnico di Milano\, Milan\, Italy\, in 2001 and 2005\, respect ively. From 2006 to 2017\, he was a faculty member of the Electrical and C omputer Engineering (ECE) Department at New Jersey Institute of Technology (NJIT)\, where he was affiliated with the Center for Wireless Information Processing (CWiP). His research interests include information theory\, ma chine learning\, wireless communications\, and neuromorphic computing. Dr Simeone is a co-recipient of the 2019 IEEE Communication Society Best Tuto rial Paper Award\, the 2018 IEEE Signal Processing Best Paper Award\, the 2017 JCN Best Paper Award\, the 2015 IEEE Communication Society Best Tutor ial Paper Award and of the Best Paper Awards of IEEE SPAWC 2007 and IEEE W RECOM 2007. He was awarded a Consolidator grant by the European Research C ouncil (ERC) in 2016. His research has been supported by the U.S. NSF\, th e ERC\, the Vienna Science and Technology Fund\, as well as by a number of industrial collaborations. He currently serves in the editorial board of the IEEE Signal Processing Magazine and is the vice-chair of the Signal Pr ocessing for Communications and Networking Technical Committee of the IEEE Signal Processing Society. He was a Distinguished Lecturer of the IEEE In formation Theory Society in 2017 and 2018. Dr Simeone is a co-author of tw o monographs\, two edited books published by Cambridge University Press\, and more than one hundred research journal papers. He is a Fellow of the I ET and of the IEEE.\n\nSpeaker(s): Prof. Osvaldo Simeone \, \n\nKingston\, Ontario\, Canada\, Virtual: https://events.vtools.ieee.org/m/274966 LOCATION:Kingston\, Ontario\, Canada\, Virtual: https://events.vtools.ieee. org/m/274966 ORGANIZER:chan-f@rmc.ca SEQUENCE:4 SUMMARY:IEEE VDL: Learning to learn to communicate URL;VALUE=URI:https://events.vtools.ieee.org/m/274966 X-ALT-DESC:Description: <br /><p><strong>Abstract</strong></p>\n<p>The appl ication of supervised learning techniques for the design of the physical l ayer of a communication link is often impaired by the limited amount of pi lot data available for each device\; while the use of unsupervised learnin g is typically limited by the need to carry out a large number of training iterations. In this talk\, meta-learning\, or learning-to-learn\, is intr oduced as a tool to alleviate these problems. The talk will consider an In ternet-of-Things (IoT) scenario in which devices transmit sporadically usi ng short packets with few pilot symbols over a fading channel. The number of pilots is generally insufficient to obtain an accurate estimate of the end-to-end channel\, which includes the effects of fading and of the trans mission-side distortion. To tackle this problem\, pilots from previous IoT transmissions are used as meta-training data in order to train a demodula tor that is able to quickly adapt to new end-to-end channel conditions fro m few pilots. Various state-of-the-art meta-learning schemes are adapted t o the problem at hand and evaluated\, including MAML\, FOMAML\, REPTILE\, and CAVIA. Both offline and online solutions are developed.</p>\n<p><stron g>Biography</strong></p>\n<p>Osvaldo Simeone is a Professor of Information Engineering with the Centre for Telecommunications Research at the Depart ment of Engineering of King's College London\, where he directs the King's Communications\, Learning and Information Processing lab. He received an M.Sc. degree (with honors) and a Ph.D. degree in information engineering f rom Politecnico di Milano\, Milan\, Italy\, in 2001 and 2005\, respectivel y. From 2006 to 2017\, he was a faculty member of the Electrical and Compu ter Engineering (ECE) Department at New Jersey Institute of Technology (NJ IT)\, where he was affiliated with the Center for Wireless Information Pro cessing (CWiP). His research interests include information theory\, machin e learning\, wireless communications\, and neuromorphic computing. Dr Sime one is a co-recipient of the 2019 IEEE Communication Society Best Tutorial Paper Award\, the 2018 IEEE Signal Processing Best Paper Award\, the 2017 JCN Best Paper Award\, the 2015 IEEE Communication Society Best Tutorial Paper Award and of the Best Paper Awards of IEEE SPAWC 2007 and IEEE WRECO M 2007. He was awarded a Consolidator grant by the European Research Counc il (ERC) in 2016. His research has been supported by the U.S. NSF\, the ER C\, the Vienna Science and Technology Fund\, as well as by a number of ind ustrial collaborations. He currently serves in the editorial board of the IEEE Signal Processing Magazine and is the vice-chair of the Signal Proces sing for Communications and Networking Technical Committee of the IEEE Sig nal Processing Society. He was a Distinguished Lecturer of the IEEE Inform ation Theory Society in 2017 and 2018. Dr Simeone is a co-author of two mo nographs\, two edited books published by Cambridge University Press\, and more than one hundred research journal papers. He is a Fellow of the IET a nd of the IEEE.</p> END:VEVENT END:VCALENDAR