Photonic tensor core and nonvolatile memory for machine learning
Volker J. Sorger · 2021
Here, we introduce an integrated photonics-based tensor core unit by strategically utilizing i) photonic parallelism via wavelength division multiplexing, ii) high Peta-operations-per second throughputs enabled by 10’s of picosecond-short delays from optoelectronics and compact photonic integrated circuitry, and iii) near-zero static power-consuming novel photonic multi-state memories based on phase-change materials featuring vanishing losses in the amorphous state. Combining these physical synergies of material, function, and system, we discuss a design and performance of a 4-bit photonic tensor core unit.