We are seeking a highly skilled and motivated candidate to investigate and implement novel concepts of neuromorphic computation, relying on the adjunct top intrinsic properties of electronics and photonics, for ultra-low latency and energy-efficient computation. This PhD research will enable a whole new class of reconfigurable electro-photonic networks, leading to highly efficient parallel processing engines.Research challenges
You will be working within the newly granted NWA ORC (Research along Routes by Consortia) 'NL-ECO' project. See here for more information: Big consortium starts research into energy-efficient information technology (utwente.nl)
. Within this extensive project framework, we will focus on combining photonic integrated technology and electronics for frontier neural network research and accelerators. The project is among the first to bridge electronic ICs and photonic ICs with the target of the co-design of both for energy efficiency, high bandwidth and low latency computation. This research will enable a whole new class of reconfigurable neural network layouts, which could lead to highly efficient optical processing engines.
Several on-chip computing paradigms have already been proposed using electrons and light. However, optimal solutions and scalability at the small energy and low latency are still far ahead. A different approach needs to be explored and realized within this program. The Photonic Neural Network Lab within the ECO group at TU/e aims to synergically collaborate with the Neuromorphic Edge Computing Systems Lab within the ES group to powerfully interface both solutions and leverage the intrinsic strengths of both electronic and photonic technologies.
The PhD student will explore the different brain-inspired neural network models and investigate suitability and mapping into an electro-photonic accelerator. He/She will address brain-inspired architecture with high dimensionality and for a complete set of neural network architectures. Depending on the best suitable implementation, dimensionality expansion, matrix multiplications, filtering, and data interpretation will be assigned to the photonic and/or electronic domains. Software-defined accelerators will be implemented to optimize data conversion and memory utilization. Image processing cases related to classifying high-resolution images and extreme hi-data-rate processing will be addressed.
The PhD work will include a mixture of activities, from neural models to physical properties of the electronic and photonic engine, from close system integration and model mapping on a chip to implementation and soft-defined programmability. This research will enable a whole new class of reconfigurable electro-photonic networks, leading to highly efficient parallel processing engines and high-impact journals.The team
The PhD position is based within the Electro-Optical Communication Systems (ECO) group and the Electronic Systems (ES) group, which are both part of the Eindhoven Hendrik Casimir Institute (EHCI) and of the Eindhoven AI System Institute (EAISI). The ECO group has about 73 members, 55 of which are PhD students and PDs, and 18 are staff members, focusing on photonic technology for communication and computation. The ES group has about 70 members, 40 of which are PhD students and focuses on ground-breaking designs for high-quality, cost-effective electronic systems for applications, including edge artificial intelligence and high-tech systems.
The EHCI has five dynamic and ambitious research groups, which are closely cooperating: a systems group, a photonic integration technology group and three materials research groups, with a predominant focus on neuromorphic and quantum computing is the ideal ecosystem where to develop this research. The EAISI brings together all AI activities of the TU/e. Top researchers from various departments and research groups work together to create new and exciting AI applications directly impacting the real world in collaboration with representatives from the industry.
The PhD student will also closely be connected to the NL-ECO consortium for idea cross-fertilization and for developing novel synergic research within the same consortium [Big consortium starts research into energy-efficient information technology (utwente.nl)
]. The student will also closely interact with other 2 PhD students who will do adjacent and connected work on the algorithms and photonic integrated sides.