Work

Hyperdimensional Photonic AI Computing

Year: 2026

Type: article

Abstract: We discuss our work on hyperdimensional photonic AI accelerators, that leverage time-space- and wavelength-division multiplexing towards enhancing compute power and energy efficiency. We benchmark their performance against state-of-the-art GPUs, demonstrating superior throughput and low-energy operation

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.24102407

Authors: Tsakyridis, Apostolos, Moralis - Pegios, Miltiadis, Kovaios, Stefanos, Prapas, Antonis, Moschos, Thodoris, Pappas, Christos, Pleros, Nikos

Language: English

Topic: Neural Networks and Reservoir Computing

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7225732175

FWCI:

Cites:

Cited by:

Related to: