I am an MRes student in Computational Statistics and Machine Learning at University College London advised by Tim Lillicrap and Karl Friston. I am also a student researcher with Vincent Vanhoucke at Google Brain’s robotics team.

My goal is to build intelligent machines based on concepts of the human brain, and evaluate them in complex simulations. Currently, I’m especially interested in model based exploration, Bayesian reinforcement learning, and temporal abstraction.

If you have any questions, feel free to contact me at [email protected].

News

2018-05-18 4 papers submitted to NIPS.

2018-04-12 Minitaur paper accepted at RSS 2018.

2018-02-16 Guest lecture at Stanford CS 20 on Variational Inference in TF.

2018-02-14 Talk at IBM Research AI on ThalNet.

Selected research

Please see my Scholar profile for a full list including article links.

• 
  Jul 14, 2018  (NIPS 2018, oral 0.6%)

  Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion

  J Buckman, D Hafner, G Tucker, E Brevdo, H Lee
• 
  Jul 14, 2018  (LLARLA 2018)

  Learning Unsupervised Latent Dynamics Models for Multi-task Continuous Control from Pixels

  D Hafner, I Fischer, T Lillicrap, D Ha, J Davidson, H Lee
• 
  Jul 14, 2018  (TADGM 2018)

  Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors

  D Hafner, D Tran, A Irpan, T Lillicrap, J Davidson
• 
  Apr 27, 2018  (RSS 2018)

  Sim-to-Real: Learning Agile Locomotion For Quadruped Robots

  J Tan, T Zhang, E Coumans, A Iscen, Y Bai, D Hafner, S Bohez, V Vanhoucke
• 
  Nov 8, 2017  (ArXiv)

  TensorFlow Agents: Efficient Batched Reinforcement Learning in TensorFlow

  D Hafner, J Davidson, V Vanhoucke
• 
  Jul 18, 2017  (NIPS 2017)

  Learning Hierarchical Information Flow with Recurrent Neural Modules

  D Hafner, A Irpan, J Davidson, N Heess
• 
  Aug 22, 2016  (ESA 2016)

  Probabilistic Routing for On-Street Parking Search

  T Arndt, D Hafner, T Kellermeier, S Krogmann, A Razmjou, M Krejca, R Rothenberger, T Friedrich
• 
  Jul 20, 2016  (ArXiv)

  Generative Interest Estimation for Document Recommendations

  D Hafner, A Immer, W Raschkowski, F Windheuser
• 
  Jul 15, 2016  (BSc thesis)

  Deep Reinforcement Learning From Raw Pixels in Doom

  D Hafner

Short biography

Danijar Hafner is a graduate student at University College London under the supervision of Tim Lillicrap and Karl Friston. He is also an intern with the robotics group at Google Brain, where he developed the brain-inspired sequence model ThalNet and the reinforcement learning library TensorFlow Agents. Danijar Hafner co-authored the book “TensorFlow for Machine Intelligence” and guest lectured in Stanford’s course “TensorFlow for Deep Learning Research”. He completed his bachelor’s thesis on deep reinforcement learning for the video game Doom at the Hasso Plattner Institute, Germany.

Links

Twitter · Github · Quora · LinkedIn · Google AI · ORCID