Job Detail

Ref. 2018-42

Project description

New computing paradigms bring the promise to extend the functionality of today’s Von Neumann based systems. Artificial intelligence and neural network based systems have proven to be good choices for the classification of unstructured data, image and speech recognition. To fully exploit the performance of such systems, dedicated hardware solutions will be required to accelerate neural network training and for fast and power efficient execution.

At IBM Research – Zurich we are establishing novel hardware technologies for neural networks by building on existing integrated electrical and optical platforms. Neural network simulations and system-level experiments complement the hardware-related activities and demonstrate the impact of the new technology platforms.

In the Neuromorphic Devices and Systems group, a post-doc position is available embedded in the SNF project NAPRECO, a collaboration with the University of St. Gallen (HSG). The main responsibility is to develop novel recurrent neural networks in the optical domain, benchmark such newly developed systems, and determine their applicability to real-world problems. The work will be at the interface between physical simulations and photonic characterization, mathematics of reservoir computing, and computer science.

Important tasks to be performed are:

• Design of next-generation integrated photonic recurrent neural networks, interfacing with collaborators in the field of mathematics.


• Developing an experimental setup to execute software-controlled training and interference tasks on the photonic hardware circuits.


• Characterization and benchmarking of the recurrent neural networks fabricated in IBM cleanroom facilities.


• Evaluation of newly developed hardware networks with common cognitive benchmarks and analysis of their applicability to real-world problems.

Duration

This post-doc position is for a duration of 18 months.

Requirements

Candidates with a strong background and interest in neural network-based systems, simulation of neural networks, and/or reservoir computing systems are encouraged to send their CV including publication list and references. Experience in the design and/or the high-speed characterization of optical systems is a plus.

Diversity

IBM is committed to diversity at the workplace. With us you will find an open, multicultural environment. Excellent flexible working arrangements enable both women and men to strike the desired balance between their professional development and their personal lives.

How to apply

For more information please contact Dr. Bert Offrein at the address indicated below. Applications should be sent by email to Dr. Bert Jan Offrein, preferably before 20 December 2018.

Dr. Bert Jan Offrein
Manager Neuromorphic Devices and Systems
Phone: +41 44 724 8572
e-mail: ofb@zurich.ibm.com