Overview

Website maintenance

I’m halfway through updating my website. Currently, recent publications are not included on this website. For now, find an overview of my publications on Google scholar. Do let me know if you notice other issues. 

Bio

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.

Recent news

  • 2-day graduate course on “adaptive hybrid intelligence” (8/28/2023)

    We are organizing a course for PhD students in the Netherlands around training RL agents to assist or collaborate with people. I’m very happy with our line-up. Read more!

  • Postdoc positions on AI for sustainable molecules and materials (4/6/2023)

    The university of Amsterdam is looking for 4 postdocs on the topic of AI for meta-materials. One of the positions will be co-advised by me, on the topic of “Machine Learning-based models of plant protein mixtures for sustainable food design”. Two other positions are co-advised by AMLab colleagues. Deadline for applications is 16 May. Apply is only possible through the form here.
    (Note that another postdoc position at the VU on learning and reasoning for medical decision making is also still open, see below).

  • TMLR paper accepted (4/3/2023)

    David Kuric’s paper on learning re-useable options was accepted to TMLR. Paper, code, and video are available on OpenReview. It proposes a gradient-based meta learning approach to discover sub-policies that are useful for rapid adaptation to different MDPs in a family of tasks. Congrats, David!

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

Highlighted publications
Gagrani, Mukul; Rainone, Corrado; Yang, Yang; Teague, Harris; Jeon, Wonseok; Hoof, Herke; Zeng, Weiliang Will; Zappi, Piero; Lott, Christopher; Bondesan, Roberto: Neural Topological Ordering for Computation Graphs. In: Advances in Neural Information Processing Systems, . (Type: Proceedings Article | Links | BibTeX)
Pol, Elise; Hoof, Herke; Oliehoek, Frans; Welling, Max: Multi-Agent MDP Homomorphic Networks. In: Proceedings of the International Conference on Learning Representations, . (Type: Proceedings Article | Links | BibTeX)
Höpner, Niklas; Tiddi, Ilaria; Hoof, Herke: Leveraging class abstraction for commonsense reinforcement learning via residual policy gradient methods. In: International Joint Conference on Artificial Intelligence, . (Type: Proceedings Article | Links | BibTeX)
Kool, Wouter; Hoof, Herke; Welling, Max: Estimating Gradients for Discrete Random Variables by Sampling without Replacement. In: International Conference on Learning Representations, 2020. (Type: Proceedings Article | Links | BibTeX)
Smith, M.; Hoof, H.; Pineau, J.: An Inference-Based Policy Gradient Method for Learning Options. In: International Conference on Machine Learning, pp. 4703-4712, 2018. (Type: Proceedings Article | Links | BibTeX)
Hoof, H. Van; Neumann, G.; Peters, J.: Non-parametric Policy Search with Limited Information Loss. In: Journal of Machine Learning Research, vol. 18, no. 73, pp. 1-46, 2017. (Type: Journal Article | Links | BibTeX)
Wang, Qi; Federici, Marco; Hoof, Herke: Bridge the Inference Gaps of Neural Processes via Expectation Maximization. In: International Conference on Learning Representations, Forthcoming. (Type: Proceedings Article | Links | BibTeX)