Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
SYNC: Symposium Advanced neuromorphic computing hardware: Towards efficient machine learning
SYNC 1: Advanced neuromorphic computing hardware: Towards efficient machine learning
SYNC 1.4: Hauptvortrag
Montag, 16. März 2020, 11:15–11:45, HSZ 01
Photonic Recurrent Ising Sampler — •Charles Roques-Carmes1, Yichen Shen1, Cristian Zanoci1, Mihika Prabhu1, Fadi Atieh1, Li Jing1, Tena Dubček1, Chenkai Mao1, Miles Johnson1, Vladimir Čeperić1, John Joannopoulos1,2, Dirk Englund1, and Marin Soljačić1 — 1Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA — 2Institute for Soldier Nanotechnologies, 500 Technology Square, Cambridge, MA 02139, USA
The inability of conventional electronic architectures to efficiently solve large combinatorial problems motivates the development of novel computational hardware. There has been much effort toward developing application-specific hardware across many different fields of engineering, such as integrated circuits, memristors, and photonics. However, unleashing the potential of such architectures requires the development of algorithms which optimally exploit their fundamental properties. We present the Photonic Recurrent Ising Sampler (PRIS), a heuristic method tailored for parallel architectures allowing fast and efficient sampling from distributions of arbitrary Ising problems. Since the PRIS relies on vector-to-fixed matrix multiplications, we suggest the implementation of the PRIS in photonic parallel networks, which realize these operations at an unprecedented speed. The PRIS provides sample solutions to the ground state of Ising models, by converging in probability to their associated Gibbs distribution. Our work suggests speedups in heuristic methods via photonic implementations of the PRIS.