BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Israel BEGIN:DAYLIGHT DTSTART:20210326T030000 TZOFFSETFROM:+0200 TZOFFSETTO:+0300 RRULE:FREQ=YEARLY;BYDAY=-1FR;BYMONTH=3 TZNAME:IDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20201025T010000 TZOFFSETFROM:+0300 TZOFFSETTO:+0200 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:IST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20211230T130052Z UID:AB7062DE-90C7-4205-B37F-D202353327A3 DTSTART;TZID=Israel:20210105T150000 DTEND;TZID=Israel:20210105T163000 DESCRIPTION:In recent years\, many remarkable achievements have been made i n the field of machine learning. While most of the initial successes were related to image\, speech and language recognition\, a recent important de velopment has been the application of these techniques to other areas. In particular\, communications systems can benefit from applying these techni ques. For example\, algorithms such as Monte Carlo Markov Chain and Monte Carlo Tree Search have been successfully used in the design of MIMO (i.e.\ , multiple antenna) transceivers. In addition\, highly quantized implement ations\, such as binarized networks\, have led to implementations that are well-suited to power-limited mobile platforms. In addition\, metaheuristi c optimization techniques such the genetic algorithm and others have been used to automatically find highly efficient deep learning architectures\, eliminating the need for lengthy and tedious manual experimentation. This lecture will describe these approaches and present some recent design exam ples. Relationships between the algorithms will be emphasized\, and import ant computational issues will be highlighted. Finally\, opportunities for future research in these areas will be suggested.\n\nGerald Sobelman is a Professor in the Department of Electrical and Computer Engineering at the University of Minnesota\, and he has served as the Director of Graduate St udies for the Graduate Program in Computer Engineering at the University o f Minnesota. He received a B.S. degree in physics from the University of C alifornia\, Los Angeles\, and M.S. and Ph.D. degrees in physics from Harva rd University. He has been a postdoctoral researcher at The Rockefeller Un iversity\, and he has held senior engineering positions at Sperry Corporat ion and Control Data Corporation.\n\nProf. Sobelman is currently a Disting uished Lecturer of the IEEE Circuits and Systems Society. He has been a me mber of the technical program committees for several IEEE conferences. He was Chair of the Technical Committee on Circuits and Systems for Communica tions of the IEEE Circuits and Systems Society\, and he has also served as an Associate Editor for IEEE Transactions on Circuits and Systems I and f or IEEE Signal Processing Letters. In addition\, he has chaired sessions a t international conferences in the areas of communications and VLSI archit ectures.\n\nProf. Sobelman has presented short courses at a number of indu strial and academic sites. He has authored or co-authored more than 150 te chnical papers and 1 book\, and he holds 12 U.S. patents.\n\nImportant: Th e participation is free of charge\, but registration is required [/registr ation-gerald-sobelman/](https://acrc.net.technion.ac.il/registration-geral d-sobelman/).\n\nVirtual: https://events.vtools.ieee.org/m/297825 LOCATION:Virtual: https://events.vtools.ieee.org/m/297825 ORGANIZER:shahar@ee.technion.ac.il SEQUENCE:0 SUMMARY:Machine Learning and Optimization for Communications and Deep Netwo rks-Prof. Gerald Sobelman\, University of Minnesota\, USA URL;VALUE=URI:https://events.vtools.ieee.org/m/297825 X-ALT-DESC:Description: <br /><p>In recent years\, many remarkable achievem ents have been made in the field of machine learning. While most of the in itial successes were related to image\, speech and language recognition\, a recent important development has been the application of these technique s to other areas. In particular\, communications systems can benefit from applying these techniques. For example\, algorithms such as Monte Carlo Ma rkov Chain and Monte Carlo Tree Search have been successfully used in the design of MIMO (i.e.\, multiple antenna) transceivers. In addition\, highl y quantized implementations\, such as binarized networks\, have led to imp lementations that are well-suited to power-limited mobile platforms. In ad dition\, metaheuristic optimization techniques such the genetic algorithm and others have been used to automatically find highly efficient deep lear ning architectures\, eliminating the need for lengthy and tedious manual e xperimentation. This lecture will describe these approaches and present so me recent design examples. Relationships between the algorithms will be em phasized\, and important computational issues will be highlighted. Finally \, opportunities for future research in these areas will be suggested.</p> \n<p><strong>Gerald Sobelman</strong>&nbsp\;is a Professor in the Departme nt of Electrical and Computer Engineering at the University of Minnesota\, and he has served as the Director of Graduate Studies for the Graduate Pr ogram in Computer Engineering at the University of Minnesota. He received a B.S. degree in physics from the University of California\, Los Angeles\, and M.S. and Ph.D. degrees in physics from Harvard University. He has bee n a postdoctoral researcher at The Rockefeller University\, and he has hel d senior engineering positions at Sperry Corporation and Control Data Corp oration.</p>\n<p>Prof. Sobelman is currently a Distinguished Lecturer of t he IEEE Circuits and Systems Society. He has been a member of the technica l program committees for several IEEE conferences. He was Chair of the Tec hnical Committee on Circuits and Systems for Communications of the IEEE Ci rcuits and Systems Society\, and he has also served as an Associate Editor for IEEE Transactions on Circuits and Systems I and for IEEE Signal Proce ssing Letters. In addition\, he has chaired sessions at international conf erences in the areas of communications and VLSI architectures.</p>\n<p>Pro f. Sobelman has presented short courses at a number of industrial and acad emic sites. He has authored or co-authored more than 150 technical papers and 1 book\, and he holds 12 U.S. patents.</p>\n<p><strong>Important</stro ng>: The participation is free of charge\, but registration is required&nb sp\;<strong><a href="https://acrc.net.technion.ac.il/registration-gerald-s obelman/">/registration-gerald-sobelman/</a>.</strong></p> END:VEVENT END:VCALENDAR