Neuromorphic Polariton Accelerator

Introduction

Exciton-polaritons, hybrid light-matter particles, have recently come into the spotlight for their peculiar properties (sizable interaction, small mass, long coherence, etc.) leading to spectacular effects such as phase transitions, superfluidity, bistability, ultra-efficient four-wave mixing, and quantum blockade.

Applications of Polaritons

On the other hand, polaritons have also been proposed for different kinds of devices, including:

1. Optical switches
2. Transistors
3. Low threshold lasers
4. Simulators

Beautiful experiments have shown proofs-of-principle for these applications. However, it is only recently that polaritons have been operating efficiently at room temperature, giving the promise of a real technological impact in the future.

Recent Developments

In a recent work made by some of the theoretical and experimental partners of this proposal, we demonstrated that such a hybrid state of matter, when used for realizing artificial neural networks, shows extremely interesting performances in terms of speed and success rate.

Project Goal

Given the strong interest in the realization of hardware-based (not simulated) artificial neural networks, the goal of PolArt is to demonstrate a new way to build artificial intelligence-dedicated circuits using polariton neural networks as optical accelerators.

Impact on Neuromorphic Computation

Thanks to this new concept device, complex applications related to neural-like processing will be efficiently implemented. This will enable neuromorphic computation to be done in small devices that cannot rely on remote, large bandwidth connections.

Interdisciplinary Collaboration

This proposal benefits from the contribution of several complementary partners coming from many different research areas, including:

• Material science
• Physics
• Optics
• Chemistry
• Genetics

Additionally, industrial participants assure the interdisciplinarity and technology-oriented target of the project.