Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Katja Hofmann

Katja Hofmann

"It's Unwieldy and It Takes a Lot of Time." Challenges and Opportunities for Creating Agents in Commercial Games

Sep 01, 2020
Mikhail Jacob, Sam Devlin, Katja Hofmann

Figure 1 for "It's Unwieldy and It Takes a Lot of Time." Challenges and Opportunities for Creating Agents in Commercial Games

Figure 2 for "It's Unwieldy and It Takes a Lot of Time." Challenges and Opportunities for Creating Agents in Commercial Games

Figure 3 for "It's Unwieldy and It Takes a Lot of Time." Challenges and Opportunities for Creating Agents in Commercial Games

* 7 pages, 3 figures, to be published in the 16th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-20)

Access Paper or Ask Questions

Guaranteeing Reproducibility in Deep Learning Competitions

May 12, 2020
Brandon Houghton, Stephanie Milani, Nicholay Topin, William Guss, Katja Hofmann, Diego Perez-Liebana, Manuela Veloso, Ruslan Salakhutdinov

* Accepted as a poster presentation to the 2019 NeruIPS Challenges in Machine Learning workshop (CiML)

Access Paper or Ask Questions

Recognizing Spatial Configurations of Objects with Graph Neural Networks

Apr 09, 2020
Laetitia Teodorescu, Katja Hofmann, Pierre-Yves Oudeyer

Figure 1 for Recognizing Spatial Configurations of Objects with Graph Neural Networks

Figure 2 for Recognizing Spatial Configurations of Objects with Graph Neural Networks

Figure 3 for Recognizing Spatial Configurations of Objects with Graph Neural Networks

Figure 4 for Recognizing Spatial Configurations of Objects with Graph Neural Networks

Access Paper or Ask Questions

Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning

Apr 07, 2020
Rémy Portelas, Katja Hofmann, Pierre-Yves Oudeyer

Figure 1 for Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning

Figure 2 for Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning

Figure 3 for Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning

Figure 4 for Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning

* Accepted to the ICLR 2020 workshop Beyond tabula rasa in RL (BeTR-RL)

Access Paper or Ask Questions

Automatic Curriculum Learning For Deep RL: A Short Survey

Mar 10, 2020
Rémy Portelas, Cédric Colas, Lilian Weng, Katja Hofmann, Pierre-Yves Oudeyer

Figure 1 for Automatic Curriculum Learning For Deep RL: A Short Survey

Figure 2 for Automatic Curriculum Learning For Deep RL: A Short Survey

Access Paper or Ask Questions

Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck

Oct 28, 2019
Maximilian Igl, Kamil Ciosek, Yingzhen Li, Sebastian Tschiatschek, Cheng Zhang, Sam Devlin, Katja Hofmann

Figure 1 for Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck

Figure 2 for Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck

Figure 3 for Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck

Figure 4 for Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck

* Published at Neurips 2019

Access Paper or Ask Questions

Better Exploration with Optimistic Actor-Critic

Oct 28, 2019
Kamil Ciosek, Quan Vuong, Robert Loftin, Katja Hofmann

Figure 1 for Better Exploration with Optimistic Actor-Critic

Figure 2 for Better Exploration with Optimistic Actor-Critic

Figure 3 for Better Exploration with Optimistic Actor-Critic

Figure 4 for Better Exploration with Optimistic Actor-Critic

* NeurIPS 2019
* 20 pages (including supplement)

Access Paper or Ask Questions

Variational Integrator Networks for Physically Meaningful Embeddings

Oct 21, 2019
Steindor Saemundsson, Alexander Terenin, Katja Hofmann, Marc Peter Deisenroth

Figure 1 for Variational Integrator Networks for Physically Meaningful Embeddings

Figure 2 for Variational Integrator Networks for Physically Meaningful Embeddings

Figure 3 for Variational Integrator Networks for Physically Meaningful Embeddings

Figure 4 for Variational Integrator Networks for Physically Meaningful Embeddings

Access Paper or Ask Questions