Project
In recent years, we have witnessed an explosion of artificial intelligence (AI) applications which will continue to grow over the next decade. An intelligent and digitized society will be ubiquitous, enabled by increased advances in nanoelectronics. Key drivers will be sensors interfacing with the physical world and taking appropriate action in a timely manner while operating with energy efficiency and flexibility to adapt. The vast majority of sensors receive analog inputs from the real world and generate analog signals to be processed.

However, digitizing these signals not only creates an enormous amount of raw data but also requires a lot of memory and high-power consumption. As the number of sensor-based IoTs grows, bandwidth limitations make it difficult to send everything back to a cloud rapidly enough for real-time processing and decision-making, especially for delay-sensitive applications such as driverless vehicles, robotics, or industrial manufacturing.

In this context, PHASTRAC proposes to develop a novel analog-to-information neuromorphic computing paradigm based on oscillatory neural networks (ONNs). We propose a first-of-its-kind and novel analog ONN computing architecture to seamlessly interface with sensors and process their analog data without any analog-to-digital conversion. ONNs are biologically inspired neuromorphic computing architecture, where neuron oscillatory behavior will be developed by innovative phase change VO2 material coupled with synapses to be developed by bilayer Mo/HfO2 RRAM devices. PHASTRAC will address key issues:

1) novel devices for implementing ONN architecture,
2) novel ONN architecture to allow analog sensor data processing, and
3) processing the data efficiently to take appropriate action.

This 'sensing-to-action' computing approach based on ONN technology will allow energy efficiency improvement 100x-1000x and establish a novel analog computing paradigm for improved future human-machine interactions. The PHASTRAC consortium includes some of Europe's strongest research groups and industries, covering from device fabrication, circuit, and architecture design to end-use applications. We will demonstrate a first-of-its-kind analog-to-information computing paradigm with industrial applications such as intelligent vehicle interior design and human-robotics interactions that opens the road for EU leadership in energy efficient edge computing.

Candidate
We are seeking a highly skilled and motivated candidate to tackle the following research challenges related to Novel Devices and Architectures for Neuromorphic Computing based on Oscillatory Neural Networks:

Oscillatory neural networks are a brain-inspired computing paradigm that allow for energy efficient and adaptive intelligent systems. By mimicking the human brain and nervous system, these computing architectures are excellent candidates for solving complex and large-scale associative learning problems. The objective of this thesis is the development of device models for emulating neurons and synapses. Oscillator neurons are based on phase change insulator-metal transition devices such as vanadium dioxide VO2. Coupling synapses are based on memristor devices based on novel bilayer MO/HfO2 (where MO stands for metal oxide), also referred to as MO/HfO2 resistive random-access memories (RRAM). Experimental development of these devices will be performed by the IBM Research Zurich team, a partner in the PHASTRAC project. Based on experimental data measurements, physical device models for VO2 oscillator and MO/HfO2 coupling will be developed. The interplay between the oscillators and coupling will be investigated to analyze the impact of device nonuniformities and their impact on ONN performance, reliability and energy. ONN circuit simulations will be performed to analyze stochastic noise and harmonic injection on the oscillators. ONN architecture design and design space exploration will be performed and evaluated for various applications such as associative memory, image segmentation, and solving combinatorial problems via the Ising model formulation. This thesis will be conducted in collaboration with PHASTRAC project partners.