PhD Position: Closed-Loop Optimization of Experimental Parameters for Neurotechnological Systems

Are you passionate about leveraging optimisation methods and machine learning techniques to enhance neurotechnological systems such as brain-computer interfaces? Do you have a solid foundation in mathematics, optimisation, machine learning and programming? If so, you are invited to become part of the Dutch national brain interfaces initiative (DBI2).

We invite applications for a PhD position to investigate sample-efficient optimisation strategies for experimental parameters. The position is to be filled as soon as possible.

Neurotechnological systems such as brain-computer interfaces (BCIs) allow us to record and interpret the brain activity of healthy users, patients or animal models in real time. Thus, BCIs not only allow us to study fundamental brain functions but they also provide applications for communication, for the control of devices, or to support the treatment of neurological or psychiatric diseases. As brain signals are individual, noisy and high dimensional, machine learning methods play a crucial role in extracting information about the ongoing brain state.

Parameters of an experimental protocol can strongly influence the measured brain signals, but parameters that are suitable for one participant may not be for another. This calls for individually optimised protocol parameters. Ideally, individual best parameters are determined in a closed-loop approach during a single experimental session. As the measured EEG / MEG / LFP / sEEG / ECoG signals are very noisy, either only a small number of parameter sets can be evaluated within one session, or each parameter set needs to be rated based on a very small amount of brain signals which, of course, may deliver noisy ratings.

The PhD project investigates optimisation approaches for parameters of neurotechnological applications with the goal to cope with noisy objective functions. The focus will be on how (1) experimental protocol parameters and (2) machine learning methods for the decoding of brain signals can be co-optimised. For both tasks, domain-specific regularisation approaches shall be explored.

As a PhD candidate, you will investigate novel optimisation strategies in simulations before translating them into experiments with a human participant in the loop. You will be expected to design and implement experimental protocols in Python. You will conduct non-invasive and invasive closed-loop experiments in our own EEG labs, in labs of our DBI2 partners or clinics, and train machine learning models to analyse our own data and the data of our scientific partners. You will help disseminate the results in high-impact papers and scientific journals, and at conferences and workshops.

This is a fixed-term (4 year), full-time position. You will be expected to participate in teaching activities involving Bachelor’s and Master’s degree students, which will take 10% of your working time. Throughout the project, you will receive guidance from Dr Michael Tangermann and be an integral part of the Data-Driven Neurotechnology Lab. The lab is situated within the Machine Learning and Neural Computing department and embedded in the Donders Institute.

Would you like to learn more about what it’s like to pursue a PhD at Radboud University? Visit the page about working as a PhD candidate.

Requirements:

• You are open-minded, enthusiastic and able to work in an international team.
• You have an excellent background in mathematics and machine learning, obtained as part of an excellent Master's degree in mathematics, applied mathematics, artificial intelligence, computer science, physics or a related discipline.
• You have a strong background in either optimisation methods for constrained/unconstrained objective functions that may be noisy, or in iterative active learning methods.
• You have strong Python programming skills, and you are familiar with libraries for optimisation as well as for machine learning. Knowledge of libraries for M/EEG processing is a benefit.
• You have a passion for research and an interest in neurotechnological systems such as brain-computer interfaces.
• Having worked with electrophysiological data such as EEG, familiarity with BCI protocols, and a neuroscience background are also considered benefits.
• You have experience with collaborating on larger software projects, and with tools and infrastructure such as compute clusters, version control, etc.
• You share our attitude towards open and reproducible science, which includes the publishing of well-documented code and FAIR datasets.
• You have an excellent command of English, which is the lingua franca in our international lab.

Salary Benefits:

• We will give you a temporary employment contract (1.0 FTE) of 1.5 years, after which your performance will be evaluated. If the evaluation is positive, your contract will be extended by 2.5 years (4-year contract).
• You will receive a starting salary of €2,901 gross per month based on a 38-hour working week, which will increase to €3,707 in the fourth year (salary scale P).
• You will receive an 8% holiday allowance and an 8,3% end-of-year bonus.
• We offer Dual Career Coaching. The Dual Career Coaching assists your partner via support, tools, and resources to improve their chances of independently finding employment in the Netherlands.
• You will receive extra days off. With full-time employment, you can choose between 30 or 41 days of annual leave instead of the statutory 20.

Work and science require good employment practices. This is reflected in Radboud University's primary and secondary employment conditions. You can make arrangements for the best possible work-life balance with flexible working hours, various leave arrangements and working from home. You are also able to compose part of your employment conditions yourself, for example, exchange income for extra leave days and receive a reimbursement for your sports subscription. And of course, we offer a good pension plan. You are given plenty of room and responsibility to develop your talents and realise your ambitions. Therefore, we provide various training and development schemes.

38 hours per week

Houtlaan 4