PhD Position Design of Emerging Neuromorphic Architectures Based on the Human Neocortex

The field of neuromorphic engineering aims at replicating the brain’s key organizing principles in custom silicon toward order-of-magnitude efficiency improvements compared to current processor architectures. Based on this promise, neuromorphic engineering is now included in worldwide research roadmaps, has seen a x10 increase in yearly research output over the last decade, and fuels interest from large industrial players as well as a flourishing landscape of new startups. However, the field has not yet matured and still misses an actionable framework to deliver on its promises [1].
In this PhD project, you will directly contribute to tackling this challenge by taking inspiration from key neuroscience insight. To do so, you will look into the brain's computational primitives at an abstraction level that is higher than individual neurons and synapses, one prominent example of which are the cortical minicolumns [2,3]. You will thus

• translate neuroscience observation into an emerging neuromorphic computer architecture,
• minimize its energy and area footprints through hardware-algorithm co-design,
• implement and tape out low-power digital (and/or mixed-signal) integrated circuits based on this architecture,
• investigate the deployment of this prototype in key edge-computing use cases, such as smart prostheses that can adapt to the patient.

You will thus pave the way to next-generation small- to large-scale neuromorphic architectures inspired by the human neocortex, targeting low-footprint adaptive devices such as personal health monitoring wearables and robotic agents.
[1] C. Frenkel and G. Indiveri, "Bottom-up and top-down approaches for the design of neuromorphic processing systems: Tradeoffs and synergies between natural and artificial intelligence," Proceedings of the IEEE, vol. 111, no. 6, pp. 623-652, 2023.
[2] D. P. Buxhoeveden and M. F. Casanova, "The minicolumn hypothesis in neuroscience," Brain, vol. 125, no. 5, pp. 935-951, 2002.
[3] T. Schwalger, M. Deger and W. Gerstner, "Towards a theory of cortical columns: From spiking neurons to interacting neural populations of finite size," PLoS Computational Biology, vol. 13, no. 4, p. e1005507, 2017.
About the Department of Microelectronics at TU Delft: https://microelectronics.tudelft.nl/
About the PI: https://microelectronics.tudelft.nl/People/bio.php?id=822
About the Cognitive Sensors Nodes and Systems (CogSys) lab: http://cogsys.tudelft.nl

Requirements:

We are looking for an enthusiastic Ph.D. candidate with a background in electrical engineering, computer architecture, or similar fields. Experience with designing multi-core digital systems *or* mixed-signal integrated circuits is a strong plus.
You are genuinely curious and passionate about the brain, while you enjoy learning beyond your comfort zone. Beyond your electrical engineering / computer architecture background, you have a keen interest to expand your expertise in the areas of machine learning and neuroscience (previous exposure in these areas is a plus). While you can take initiative, you are also a team player in order to make the best of interdisciplinary collaborations in a research team.
You have an outstanding motivation but are unsure whether you meet these requirements? Unconventional backgrounds are welcome; you are still encouraged to apply (see application procedure hereafter, ensure your cover letter is written accordingly and highlights how you plan to overcome any gap).
Doing a PhD at TU Delft requires English proficiency at a certain level to ensure that the candidate is able to communicate and interact well, participate in English-taught Doctoral Education courses, and write scientific articles and a final thesis. For more details please check the Graduate Schools Admission Requirements.

Salary Benefits:

Doctoral candidates will be offered a 4-year period of employment in principle, but in the form of 2 employment contracts. An initial 1,5 year contract with an official go/no go progress assessment within 15 months. Followed by an additional contract for the remaining 2,5 years assuming everything goes well and performance requirements are met.
Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities, increasing from € 2770 per month in the first year to € 3539 in the fourth year. As a PhD candidate you will be enrolled in the TU Delft Graduate School. The TU Delft Graduate School provides an inspiring research environment with an excellent team of supervisors, academic staff and a mentor. The Doctoral Education Programme is aimed at developing your transferable, discipline-related and research skills.
The TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged.

36 - 40 hours per week

Mekelweg 2