PhD or Postdoctoral Researcher in Bio-inspired, Model-Based Reinforcement Learning

alternate text


Location: Institute of Neuroinformatics
and ETH AI Center, ETH Zurich, Switzerland

Start Date: As early as June 1, 2023

The Grewe lab (grewelab.org) at the Institute of Neuroinformatics (www.ini.uzh.ch/en.html) and the ETH AI Center (ai.ethz.ch) at ETH Zurich invites applications for a Postdoctoral position in the area of bio-inspired, model-based reinforcement learning. We are particularly interested in candidates who can contribute to our ongoing research endeavors to understand and model the complex and dynamic patterns of neuronal activity inspired by how the brain operates and solves tasks.


Research Background:

Our lab's research focuses on understanding how the brain alters its internal neuronal activity patterns, which encode information across large neuronal ensembles and multiple brain areas, to facilitate learning. We are particularly interested in the mechanisms underlying changes in network information processing related to learning a model of the world.

Given the inherent complexity, unreliability, and multidimensionality of neuronal activity patterns in the brain, this is a challenging endeavor. To better understand learning in the brain we employ in in vivo brain recording methods in mice to characterize learning-induced changes in neuronal ensemble activity (RL in mice).

Simultaneously, we aim to develop biologically-inspired multi-layer artificial neuronal network models (ANNs) that mimic the information processing and storage capabilities observed in real biological networks (e.g. to solve an RL problem). We place significant emphasis on reverse-engineering neuronal network function at a very abstract level and to understand the fundamental principles that determine learning-induced changes in neuronal networks (biological and artificial).


Role Description:

As a PPhD or ostdoctoral researcher, you will be tasked with developing and implementing innovative bio-inspired, model-based reinforcement learning algorithms. Your role will involve close collaboration with a multidisciplinary team of researchers at the ETH AI Center while contributing to our research output through high-quality publications.


Key Responsibilities:

  • Design and implement bio-inspired, model-based reinforcement learning algorithms.

  • Conduct rigorous RL experiments and analysis to test and refine these algorithms.

  • Collaborate with a multidisciplinary team to advance our collective research goals.

  • Contribute to research output by producing high-quality publications.

  • Present research findings at internal and external meetings and conferences.

  • Contribute to the broader academic community through peer review and other service roles.


Essential Qualifications:

  • MSc or Ph.D. in Theoretical Neuroscience, Computer Science, Artificial Intelligence, or a closely related field (e.g. Physics).

  • Background in reinforcement learning, particularly model-based methods.

  • Familiarity with bio-inspired neuronal network approaches to machine learning.

  • Proven track record demonstrated by publications in reputable journals (PD only).

  • Excellent written and verbal communication skills in English.


Desirable Qualifications:

  • Experience with computational or theoretical neuroscience and bio-inspired deep learning methods.

  • Proficiency in programming languages commonly used in AI research (e.g., Python, R, Matlab).

  • Experience with deep learning frameworks (e.g., TensorFlow, PyTorch).


Salary:

The salary for this position will be in accordance with the standard ETH PhD/postdoc salary.

https://ethz.ch/en/the-eth-zurich/working-teaching-and-research/welcome-center/employment-contract-and-salary/salary.html


Application Process:

Interested applicants should submit a detailed CV, a cover letter explaining their interest in the position and contact information for three references to bgrewe (at) ethz.ch

ETH Zurich is committed to increasing diversity in its workforce and encourages applications from all qualified individuals, regardless of their gender, age, cultural background, or disability status.

Please note that only shortlisted candidates will be contacted for an interview.

back
back