Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
SYNC 1: Advanced neuromorphic computing hardware: Towards efficient machine learning
Montag, 16. März 2020, 09:30–12:15, HSZ 01
Recently novel computational approaches such as neural networks are revolutionizing computation. At the same time, we experience that the performance growth of digital microchips is saturating and the energy consumption of classical digital electronic processors is becoming a serious issue. This impasse has re-invigorated learning from the brain with its amazing intelligence-per-watt ratio and the exploration of unconventional physical substrates and nonlinear phenomena.
Our symposium will present the recent progress and future perspectives of neuro-inspired computing based on solid state systems and its relation to machine learning. This includes not only important aspects of novel computational architectures in unconventional substrates but also new theoretical concepts of computing in non-digital, "brain-like" physical substrates.
The chosen topic has highly interdisciplinary as we aim at bringing together researchers from material science, machine learning, computer engineering, nonlinear dynamics with exciting talks of renowned international expertise in the field.
|09:30||SYNC 1.1||Hauptvortrag: Photonic Reservoir Computing and its Application to Optical Communication — •Ingo Fischer and Apostolos Argyris|
|10:00||SYNC 1.2||Hauptvortrag: Metal-oxide resistance switching memory devices as artificial synapses for brain-inspired computing — •Sabina Spiga|
|10:30||SYNC 1.3||Hauptvortrag: Towards brain-inspired photonic computing — •Wolfram Pernice|
|11:00||15 min. break|
|11:15||SYNC 1.4||Hauptvortrag: Photonic Recurrent Ising Sampler — •Charles Roques-Carmes, Yichen Shen, Cristian Zanoci, Mihika Prabhu, Fadi Atieh, Li Jing, Tena Dubček, Chenkai Mao, Miles Johnson, Vladimir Čeperić, John Joannopoulos, Dirk Englund, and Marin Soljačić|
|11:45||SYNC 1.5||Hauptvortrag: Beyond von Neumann systems: Computational memory for efficient AI — •Irem Boybat|