Research Projects

Below please find a summary of research projects where I acted as PI, work-package leader or was the early stage researcher (Marie Skłodowska-Curie). I excluded projects in which I had no responsibility beyond PhD student or postdoc without.

ORCHID 2021 - 2024 (ANR)

More information coming soon.

POST-DIGITAL 2020-2024 (H2020, MSCA-ITN)

European Training Network on Post-Digital Computing – POSTDIGITAL is an interdisciplinary training network comprising internationally leading teams from academia, research centres and industry, including IBM, Thales and three highly reputed SMEs. POST-DIGITAL will provide a unique training opportunity to a cohort of 15 early stage researchers (ESRs) in the inter-disciplinary fields of emerging disruptive neuromorphic computational technologies and their applications. The strong industrial presence in the network will bridge the gap between early stage innovation and utilization, providing ESRs with the experience of practical applications and solutions beyond traditional digital methods. POST-DIGITAL has the ambition and the vision to create a new generation of scientific and industrial leaders that will greatly contribute to strengthening Europe’s human resources and industry competitiveness in future digital and post digital economy and technology.

More information:

ANACONDA 2020-2024 (ANR)

ANACONDA aims at developing large-scale and ultrafast photonic neuromorphic hardware based on spiking artificial neurons.  We plan to leverage photonic recurrent SNNs to design a reservoir computer (RC) [4] with spiking nodes (photonic Liquid State Machine), and ultra-fast photonic spike processing. The use of photonic neuromorphic systems is motivated by potential parallelism and ultra-high speed: excitable optical neurons have demonstrated spike durations on the order of a few hundred picoseconds and multi-GHz spike rates. This is approximately 3 and 6 orders of magnitude faster than their electronic or biological counterparts, respectively. Finally, light propagation during inter-neuron communication consumes little energy due to the small absorption and cross-talk between photonic signal channels. Photonic neuromorphic hardware is therefore potentially highly energy efficient.

More information:

NeuroQNet II 2019-2023 (Volkswagen Foundation)

NeuroQNet II will establish (i) an universal platform for nonlinear photonic networks, and (ii) a revolutionary concept for 3D integrated photonic networks.With respect to (i) we will strongly advance the capabilities of the nanophotonic hardware platform developed in NeuroQNet I. The current quantum dot micropillar arrays (QDMPA) are well suited for photonic reservoir computing (RC), yet the non-ideal spectral homogeneity and optical gain frustrate other high-impact applications like coherent annealing computing and coherent beam combining. Fundamentally, these applications are based on identical architectures as photonic RC: coupling between the individual lasers and / or optical injection. Thus, to utilize the full potential of our photonic hardware platform we will optimize it to demonstrate coherent annealing and to investigate the possibility of coherent beam combining for the first time in a coupled microlaser system.Secondly, regarding (ii) we will revolutionize neuromorphic photonic circuits through integration into 3D photonic waveguides via nanoscribe 3D printing. Such an architecture will be of extremely far-reaching impact. In the currently employed 2D-only substrates, the integration of networks is not scalable. Leveraging the scaling between 2D-emitters and 3D-connections, we will overcome this bottleneck to crucially enhance the computation power of neuromorphic systems.

Newrons, 2018 - 2022 (H2020, Marie Skłodowska-Curie MULTIPLY co-fund)

NEWRONS project essentially consists in the experimental realization of an autonomous all-optical NN, which will be implemented in the nonlinear photonics laboratory at the FEMTO-ST institute. This project inspires from the advances of the host group in implementing large fully analogue and parallel recurrent NNs to perform reservoir computing. NEWRONS will qualitative boost the state-of-the-art creating the first world- wide all-optical NN that does not require a supervising computer during operation. Thus enabling the autonomous operation necessary to build an optical processing unit (OPU). We will test and optimize the performance of our system in most important benchmarks. We will particularly focus on the proof-of-concept experiment of telecommunication signal conditioning using our autonomous all-optical concept, planned for the secondment period at the group of Prof. Sergei Turitsyn at AiPT (Aston University).

More information:

NeuroQNet 2016-2019 (Volkswagen Foundation)

The main objective (OI) of NeuroQNet is to implement Reservoir Computing (RC), a neuro-inspired information processing scheme, in an optical network of nano-structures. Quantum Dot (QD) micropillar arrays (QDMPA) (OII) will be connected in an all-optical networks including 100s of such emitters (OIII). The project is based upon an interdisciplinary approach: the Neuromorphic computation is based on Reservoir Computing, QDMPAs provide the neural hardware and diffractive coupling will form the network to implement the neuro-inspired scheme (see Figure).

The project NeuroQNet was funded under the Integration of Molecular Components in Functional Macroscopic Systems initiative of the Volkswagen Stiftung (Volkswagen Foundation). The project duration is three years, being funded from 01.01.2016 until the 31.12.2018. Principle investigators are Prof. Reitzensein at the TU-Berlin, Germany, and me here at FEMTO-ST. Within the project framework, we strongly collaborate with Prof. Ingo Fischer at the IFISC, Palma de Mallorca, Spain.

NOVALIS 2011 - 2013 (FP7, Marie Skłodowska-Curie IEF)

The Marie Curie IEF NOVALIS project (Project # 275840) aimed at realizing a Liquid-State-Machine (LSM) based on multiple semiconductor lasers serving as nonlinear nodes. The LSM represents a novel machine learning concept, also referred to as Reservoir-Computing (RC) or nonlinear transient computing. Central to the concept is the realisation of a network of nonlinear nodes, in which the information to be processed induces complex nonlinear transient dynamics. A linear weighted sum of the individual transients provide the result of the desired computation.