Rump Sessions will be held on September 21 from 17:45 to 19:15 at Tachibana Conference Hall and Hagi Conference Hall.
Artificial neural networks have stricken back against conventional algorithm-based computing in recent years, and hence R&D of neural device/hardware is massively driven by practical demands for realizing compact, low power, and brain-morphic (brain-like) intelligent artifacts. Trends in building AI systems are of course based on silicon CMOS and memory technologies, which results in architectural competition of power efficiency and memory-logic bandwidth, as in present Neumann-based computer systems engineering. Although present fundamental CMOS devices are definitely logic and memory devices, we here introduce a “virtual” unit, i.e., brain-morphic 3-D AI devices and materials, optimized for physical reconstruction of fundamental brain structures. The device aims at not only implementing conventional AI systems, especially in the cloud edge, but encouraging both emergence and growth of advanced neuromorphic computing/AI systems. Five professional panelists were invited to discuss the possibility and difficulty, as well as about what these devices bring to us, towards launching grand challenges for possible R&D of the AI-specific novel fundamental devices.
Spintronics utilizes the spin degree of freedom of electrons in addition to the charge, which can expand the capabilities of electronic and optical devices. Discovery of giant magnetoresistance (GMR) and tunneling magnetoresistance (TMR) have been used to detect magnetic field, which boosted the areal density of hard disk drive and made Magnetic random access memories (MRAMs) dense. Spin-transfer-torque is used to write data in denser MRAMs having the most durable non-volatile memory.
Now, evolutions and new spintronics physics are coming up one after another. The sensitivity of magnetic tunnel junctions (MTJs) becomes very high to detect very tiny magnetic field from our heart at room temperature. Spin-orbit-torque writing, especially spin-Hall effect is expected to have much higher write-efficiency and much better endurance than conventional existing STT writing does. Voltage-change-magnetic-anisotropy (VCMA) effect is expected to extremely reduce energy consumption in a memory- hierachy. 100 % circularly polarized light can be emitted at room temperature by a spin-LED.
This session will introduce ongoing researches and developments on spintronics and the possibility of the innovations will be discussed.