Herke van Hoof

Herke van Hoof is currently assistant professor at the University of Amsterdam in the Netherlands, where he is part of the Amlab. He is interested in reinforcement learning with structured data and prior knowledge. Reinforcement learning is a very general framework, but this tends to result in extremely data-hungry algorithms. Exploiting structured prior knowledge, or using value function or policy parametrizations that respect known structural properties, is a promising avenue to learn more with less data. Examples of this line of work include reinforcement learning (RL) for combinatorial optimisation, RL with symbolic prior knowledge, and equivariant RL.  

Before joining the University of Amsterdam, Herke van Hoof was a postdoc at McGill University in Montreal, Canada, where he worked with Professors Joelle Pineau, Dave Meger, and Gregory Dudek. He obtained his PhD at TU Darmstadt, Germany, under the supervision of Professor Jan Peters, where he graduated in November 2016. Herke got his bachelor and master degrees in Artificial Intelligence at the University of Groningen in the Netherlands.

• ICLR paper accepted (1/23/2023)

  Qi and Marco’s paper got accept for presentation at ICLR. Congratulations! The submitted version is available here, camera-ready version is coming soon.

• Neurips papers (11/29/2022)

  This week, Qi Wang is presenting his paper on meta-learning mixture-of-expert neural processes and Mukul Gagrani and Corrado Rainone their paper on neural topological ordering for computation graphs at NeurIPS. Furthermore, Matthew Macfarlane will present preliminary work on learning tree-based policies in the Deep Reinforcement Learning workshop.

• ICML paper accepted (5/16/2022)

  Qi’s ICML paper on model-based meta learning got accepted. Congratulation, Qi! A pre-print is available on here.

An archive of news items can be found on the News page.

A full list of publications can be found at the Publications page.