BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:US/Eastern BEGIN:DAYLIGHT DTSTART:20220313T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20221106T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20220918T120010Z UID:AF5581AF-0013-4BC9-BB6E-A3FBCC07545B DTSTART;TZID=US/Eastern:20220914T110000 DTEND;TZID=US/Eastern:20220914T120000 DESCRIPTION:FREE Webinar\n\nMagnetic tunnel junctions (MTJs) represent the most useful devices coming out of spintronics research. Besides serving as the key components for magnetic field sensors and digital magnetic random access memories\, magnetic tunnel junctions have recently been studied as building blocks for non-conventional computing\, via utilizing their func tions of non-linearity\, stochasticity\, etc.\n\nIn this presentation we b egin with the fundamental aspects of these devices and then share some of our research\, as well as reliability-related issues found.\n\nIn one of t he early works\, we demonstrated that by designing multi-domain MTJs\, one can realize synaptic devices and activation function devices for convolut ional neural network\, where the synaptic weight and threshold function ar e realized by controlling the position of magnetic domain walls. Recently\ , we explore the possibility of building Hopfield neural network with MTJs by using their oscillatory or probabilistic switching properties. These c omputing hardwares\, known as Boltzmann machine or Ising machine can be us ed to solve NP-hard combinatorial optimization problems more efficiently t han traditional von Neumann architectures. Particularly\, we look into the dynamical behavior of an electrically coupled array of gigahertz spin Hal l nano-oscillators\, a device where the magnetic layers of the forming MTJ s undergo persistent precession. By developing a general analytical framew ork that describes injection locking of spin Hall oscillators with large p recession angles\, we show the mapping between the coupled oscillators’ properties and the Ising model. We then integrate the analytical model int o a versatile Verilog-A device that can emulate the coupled dynamics of sp in Hall oscillators in circuit simulators. This abstract model allows for the analysis of the performance of the spin Hall oscillator network at the circuit level using conventional electronic components and considering ph ase noise and scalability. The results provide design insights and analysi s tools toward the realization of a CMOS-integrated spin Hall oscillator I sing machine operating with a high degree of time\, space\, and energy eff iciency.\n\nSpeaker(s): Luqiao Liu\, \n\nAgenda: \n11:00 AM Technical Pres entation\n\n11:45 AM Questions and Answers\n\n12:00 PM Adjournment\n\nVirt ual: https://events.vtools.ieee.org/m/322887 LOCATION:Virtual: https://events.vtools.ieee.org/m/322887 ORGANIZER:michael.bannan@ieee.org SEQUENCE:10 SUMMARY:Webinar - Magnetic Tunnel Junctions for Non-Conventional Computing URL;VALUE=URI:https://events.vtools.ieee.org/m/322887 X-ALT-DESC:Description: <br /><p><strong><em>FREE Webinar</em></strong></p> \n<p class="western"><span style="font-family: Times New Roman\, serif\;"> Magnetic tunnel junctions (MTJs) represent the most useful devices coming out of spintronics research. Besides serving as the key components for mag netic field sensors and digital magnetic random access memories\, magnetic tunnel junctions have recently been studied as building blocks for non-co nventional computing\, via utilizing their functions of non-linearity\, st ochasticity\, etc. </span></p>\n<p class="western"><span style="font-famil y: Times New Roman\, serif\;">In this presentation we begin with the funda mental aspects of these devices and then share some of our research\, as w ell as reliability-related issues found. </span></p>\n<p class="western">< span style="font-family: Times New Roman\, serif\;">In one of the early wo rks\, we demonstrated that by designing multi-domain MTJs\, one can realiz e synaptic devices and activation function devices for convolutional neura l network\, where the synaptic weight and threshold function are realized by controlling the position of magnetic domain walls. Recently\, we explor e the possibility of building Hopfield neural network with MTJs by using t heir oscillatory or probabilistic switching properties. These computing ha rdwares\, known as Boltzmann machine or Ising machine can be used to solve NP-hard combinatorial optimization problems more efficiently than traditi onal von Neumann architectures. Particularly\, we look into the dynamical behavior of an electrically coupled array of gigahertz spin Hall nano-osci llators\, a device where the magnetic layers of the forming MTJs undergo p ersistent precession. By developing a general analytical framework that de scribes injection locking of spin Hall oscillators with large precession a ngles\, we show the mapping between the coupled oscillators&rsquo\; proper ties and the Ising model. We then integrate the analytical model into a ve rsatile Verilog-A device that can emulate the coupled dynamics of spin Hal l oscillators in circuit simulators. This abstract model allows for the an alysis of the performance of the spin Hall oscillator network at the circu it level using conventional electronic components and considering phase no ise and scalability. The results provide design insights and analysis tool s toward the realization of a CMOS-integrated spin Hall oscillator Ising m achine operating with a high degree of time\, space\, and energy efficienc y.</span></p>\n<p>&nbsp\;</p><br /><br />Agenda: <br /><p>&nbsp\;</p>\n<p> <strong>11:00 AM</strong>&nbsp\;&nbsp\;&nbsp\;Technical Presentation</p>\n <p><strong>11:45 AM</strong>&nbsp\;&nbsp\; Questions and Answers</p>\n<p>< strong>12:00 PM</strong>&nbsp\;&nbsp\;&nbsp\;Adjournment</p> END:VEVENT END:VCALENDAR