Jost Tobias Springenberg

Collect & Infer -- a fresh look at data-efficient Reinforcement Learning

Aug 23, 2021
Martin Riedmiller, Jost Tobias Springenberg, Roland Hafner, Nicolas Heess

On Multi-objective Policy Optimization as a Tool for Reinforcement Learning

Jun 15, 2021
Abbas Abdolmaleki, Sandy H. Huang, Giulia Vezzani, Bobak Shahriari, Jost Tobias Springenberg, Shruti Mishra, Dhruva TB, Arunkumar Byravan, Konstantinos Bousmalis, Andras Gyorgy, Csaba Szepesvari, Raia Hadsell, Nicolas Heess, Martin Riedmiller

Figure 2 for On Multi-objective Policy Optimization as a Tool for Reinforcement Learning

Figure 3 for On Multi-objective Policy Optimization as a Tool for Reinforcement Learning

Figure 4 for On Multi-objective Policy Optimization as a Tool for Reinforcement Learning

Access Paper or Ask Questions

Rethinking Exploration for Sample-Efficient Policy Learning

Jan 23, 2021
William F. Whitney, Michael Bloesch, Jost Tobias Springenberg, Abbas Abdolmaleki, Martin Riedmiller

Figure 2 for Rethinking Exploration for Sample-Efficient Policy Learning

Figure 3 for Rethinking Exploration for Sample-Efficient Policy Learning

Figure 4 for Rethinking Exploration for Sample-Efficient Policy Learning

Access Paper or Ask Questions

Training Generative Adversarial Networks by Solving Ordinary Differential Equations

Oct 28, 2020
Chongli Qin, Yan Wu, Jost Tobias Springenberg, Andrew Brock, Jeff Donahue, Timothy P. Lillicrap, Pushmeet Kohli

Figure 2 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations

Figure 3 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations

Figure 4 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations

Access Paper or Ask Questions

Learning Dexterous Manipulation from Suboptimal Experts

Oct 16, 2020
Rae Jeong, Jost Tobias Springenberg, Jackie Kay, Daniel Zheng, Yuxiang Zhou, Alexandre Galashov, Nicolas Heess, Francesco Nori

Figure 2 for Learning Dexterous Manipulation from Suboptimal Experts

Figure 3 for Learning Dexterous Manipulation from Suboptimal Experts

Figure 4 for Learning Dexterous Manipulation from Suboptimal Experts

Access Paper or Ask Questions

Local Search for Policy Iteration in Continuous Control

Oct 12, 2020
Jost Tobias Springenberg, Nicolas Heess, Daniel Mankowitz, Josh Merel, Arunkumar Byravan, Abbas Abdolmaleki, Jackie Kay, Jonas Degrave, Julian Schrittwieser, Yuval Tassa, Jonas Buchli, Dan Belov, Martin Riedmiller

Figure 2 for Local Search for Policy Iteration in Continuous Control

Figure 3 for Local Search for Policy Iteration in Continuous Control

Figure 4 for Local Search for Policy Iteration in Continuous Control

Access Paper or Ask Questions

Critic Regularized Regression

Jun 26, 2020
Ziyu Wang, Alexander Novikov, Konrad Żołna, Jost Tobias Springenberg, Scott Reed, Bobak Shahriari, Noah Siegel, Josh Merel, Caglar Gulcehre, Nicolas Heess, Nando de Freitas

Figure 2 for Critic Regularized Regression

Figure 3 for Critic Regularized Regression

Figure 4 for Critic Regularized Regression

* 23 pages

Access Paper or Ask Questions

Simple Sensor Intentions for Exploration

May 15, 2020
Tim Hertweck, Martin Riedmiller, Michael Bloesch, Jost Tobias Springenberg, Noah Siegel, Markus Wulfmeier, Roland Hafner, Nicolas Heess

Figure 2 for Simple Sensor Intentions for Exploration

Figure 3 for Simple Sensor Intentions for Exploration

Figure 4 for Simple Sensor Intentions for Exploration

Access Paper or Ask Questions

Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning

Feb 23, 2020
Noah Y. Siegel, Jost Tobias Springenberg, Felix Berkenkamp, Abbas Abdolmaleki, Michael Neunert, Thomas Lampe, Roland Hafner, Nicolas Heess, Martin Riedmiller

Figure 2 for Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning

Figure 3 for Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning

Figure 4 for Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning

* To appear in ICLR 2020

Access Paper or Ask Questions

Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics

Jan 02, 2020
Michael Neunert, Abbas Abdolmaleki, Markus Wulfmeier, Thomas Lampe, Jost Tobias Springenberg, Roland Hafner, Francesco Romano, Jonas Buchli, Nicolas Heess, Martin Riedmiller

Figure 2 for Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics

Figure 3 for Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics

Figure 4 for Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics

* Presented at the 3rd Conference on Robot Learning (CoRL 2019), Osaka, Japan. Video: https://youtu.be/eUqQDLQXb7I

Access Paper or Ask Questions

Quinoa: a Q-function You Infer Normalized Over Actions

Nov 05, 2019
Jonas Degrave, Abbas Abdolmaleki, Jost Tobias Springenberg, Nicolas Heess, Martin Riedmiller

Figure 2 for Quinoa: a Q-function You Infer Normalized Over Actions

* Deep RL Workshop/NeurIPS

Access Paper or Ask Questions

Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models

Oct 09, 2019
Arunkumar Byravan, Jost Tobias Springenberg, Abbas Abdolmaleki, Roland Hafner, Michael Neunert, Thomas Lampe, Noah Siegel, Nicolas Heess, Martin Riedmiller

Figure 2 for Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models

Figure 3 for Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models

Figure 4 for Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models

* To appear at the 3rd annual Conference on Robot Learning, Osaka, Japan (CoRL 2019). 24 pages including appendix (main paper - 8 pages)

Access Paper or Ask Questions

V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control

Sep 26, 2019
H. Francis Song, Abbas Abdolmaleki, Jost Tobias Springenberg, Aidan Clark, Hubert Soyer, Jack W. Rae, Seb Noury, Arun Ahuja, Siqi Liu, Dhruva Tirumala, Nicolas Heess, Dan Belov, Martin Riedmiller, Matthew M. Botvinick

Figure 2 for V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control

Figure 3 for V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control

Figure 4 for V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control

* * equal contribution

Access Paper or Ask Questions

Regularized Hierarchical Policies for Compositional Transfer in Robotics

Jun 27, 2019
Markus Wulfmeier, Abbas Abdolmaleki, Roland Hafner, Jost Tobias Springenberg, Michael Neunert, Tim Hertweck, Thomas Lampe, Noah Siegel, Nicolas Heess, Martin Riedmiller

Figure 2 for Regularized Hierarchical Policies for Compositional Transfer in Robotics

Figure 3 for Regularized Hierarchical Policies for Compositional Transfer in Robotics

Figure 4 for Regularized Hierarchical Policies for Compositional Transfer in Robotics

* Preprint. Under review. Addressed typos

Access Paper or Ask Questions

Robust Reinforcement Learning for Continuous Control with Model Misspecification

Jun 18, 2019
Daniel J. Mankowitz, Nir Levine, Rae Jeong, Abbas Abdolmaleki, Jost Tobias Springenberg, Timothy Mann, Todd Hester, Martin Riedmiller

Figure 2 for Robust Reinforcement Learning for Continuous Control with Model Misspecification

Figure 3 for Robust Reinforcement Learning for Continuous Control with Model Misspecification

Figure 4 for Robust Reinforcement Learning for Continuous Control with Model Misspecification

Access Paper or Ask Questions

Self-supervised Learning of Image Embedding for Continuous Control

Jan 03, 2019
Carlos Florensa, Jonas Degrave, Nicolas Heess, Jost Tobias Springenberg, Martin Riedmiller

Figure 2 for Self-supervised Learning of Image Embedding for Continuous Control

Figure 3 for Self-supervised Learning of Image Embedding for Continuous Control

Figure 4 for Self-supervised Learning of Image Embedding for Continuous Control

* Contributed talk at Inference to Control workshop at NeurIPS2018

Access Paper or Ask Questions

Relative Entropy Regularized Policy Iteration

Dec 05, 2018
Abbas Abdolmaleki, Jost Tobias Springenberg, Jonas Degrave, Steven Bohez, Yuval Tassa, Dan Belov, Nicolas Heess, Martin Riedmiller

Figure 2 for Relative Entropy Regularized Policy Iteration

Figure 3 for Relative Entropy Regularized Policy Iteration

Figure 4 for Relative Entropy Regularized Policy Iteration

Access Paper or Ask Questions

Maximum a Posteriori Policy Optimisation

Jun 14, 2018
Abbas Abdolmaleki, Jost Tobias Springenberg, Yuval Tassa, Remi Munos, Nicolas Heess, Martin Riedmiller

Figure 2 for Maximum a Posteriori Policy Optimisation

Figure 3 for Maximum a Posteriori Policy Optimisation

Figure 4 for Maximum a Posteriori Policy Optimisation

Access Paper or Ask Questions

Deep learning with convolutional neural networks for EEG decoding and visualization

Jun 08, 2018
Robin Tibor Schirrmeister, Jost Tobias Springenberg, Lukas Dominique Josef Fiederer, Martin Glasstetter, Katharina Eggensperger, Michael Tangermann, Frank Hutter, Wolfram Burgard, Tonio Ball

Figure 2 for Deep learning with convolutional neural networks for EEG decoding and visualization

Figure 3 for Deep learning with convolutional neural networks for EEG decoding and visualization

Figure 4 for Deep learning with convolutional neural networks for EEG decoding and visualization

* A revised manuscript (with the new title) has been accepted at Human Brain Mapping, see http://onlinelibrary.wiley.com/doi/10.1002/hbm.23730/full

Access Paper or Ask Questions

Graph networks as learnable physics engines for inference and control

Jun 04, 2018
Alvaro Sanchez-Gonzalez, Nicolas Heess, Jost Tobias Springenberg, Josh Merel, Martin Riedmiller, Raia Hadsell, Peter Battaglia

Figure 2 for Graph networks as learnable physics engines for inference and control

Figure 3 for Graph networks as learnable physics engines for inference and control

* ICML 2018

Access Paper or Ask Questions

Learning by Playing - Solving Sparse Reward Tasks from Scratch

Feb 28, 2018
Martin Riedmiller, Roland Hafner, Thomas Lampe, Michael Neunert, Jonas Degrave, Tom Van de Wiele, Volodymyr Mnih, Nicolas Heess, Jost Tobias Springenberg

Figure 2 for Learning by Playing - Solving Sparse Reward Tasks from Scratch

Figure 3 for Learning by Playing - Solving Sparse Reward Tasks from Scratch

Figure 4 for Learning by Playing - Solving Sparse Reward Tasks from Scratch

* A video of the rich set of learned behaviours can be found at https://youtu.be/mPKyvocNe_M

Access Paper or Ask Questions

Learning to Generate Chairs, Tables and Cars with Convolutional Networks

Aug 02, 2017
Alexey Dosovitskiy, Jost Tobias Springenberg, Maxim Tatarchenko, Thomas Brox

Figure 2 for Learning to Generate Chairs, Tables and Cars with Convolutional Networks

Figure 3 for Learning to Generate Chairs, Tables and Cars with Convolutional Networks

Figure 4 for Learning to Generate Chairs, Tables and Cars with Convolutional Networks

* v4: final PAMI version. New architecture figure

Access Paper or Ask Questions

Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments

Jul 23, 2017
Jingwei Zhang, Jost Tobias Springenberg, Joschka Boedecker, Wolfram Burgard

Figure 2 for Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments

Figure 3 for Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments

Figure 4 for Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments

* Camera ready version for IROS 2017

Access Paper or Ask Questions

Asynchronous Stochastic Gradient MCMC with Elastic Coupling

Dec 08, 2016
Jost Tobias Springenberg, Aaron Klein, Stefan Falkner, Frank Hutter

Figure 2 for Asynchronous Stochastic Gradient MCMC with Elastic Coupling

Access Paper or Ask Questions

Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

Apr 30, 2016
Jost Tobias Springenberg

Figure 2 for Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

Figure 3 for Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

Figure 4 for Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

Access Paper or Ask Questions

Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images

Nov 20, 2015
Manuel Watter, Jost Tobias Springenberg, Joschka Boedecker, Martin Riedmiller

Figure 2 for Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images

Figure 3 for Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images

Figure 4 for Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images

* Final NIPS version

Access Paper or Ask Questions

Multimodal Deep Learning for Robust RGB-D Object Recognition

Aug 18, 2015
Andreas Eitel, Jost Tobias Springenberg, Luciano Spinello, Martin Riedmiller, Wolfram Burgard

Figure 2 for Multimodal Deep Learning for Robust RGB-D Object Recognition

Figure 3 for Multimodal Deep Learning for Robust RGB-D Object Recognition

Figure 4 for Multimodal Deep Learning for Robust RGB-D Object Recognition

* Final version submitted to IROS'2015, results unchanged, reformulation of some text passages in abstract and introduction

Access Paper or Ask Questions

Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Jun 19, 2015
Alexey Dosovitskiy, Philipp Fischer, Jost Tobias Springenberg, Martin Riedmiller, Thomas Brox

Figure 2 for Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Figure 3 for Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Figure 4 for Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

* PAMI submission. Includes matching experiments as in arXiv:1405.5769v1. Also includes new network architectures, experiments on Caltech-256, experiment on combining Exemplar-CNN with clustering

Access Paper or Ask Questions

Striving for Simplicity: The All Convolutional Net

Apr 13, 2015
Jost Tobias Springenberg, Alexey Dosovitskiy, Thomas Brox, Martin Riedmiller

Figure 2 for Striving for Simplicity: The All Convolutional Net

Figure 3 for Striving for Simplicity: The All Convolutional Net

Figure 4 for Striving for Simplicity: The All Convolutional Net

* accepted to ICLR-2015 workshop track; no changes other than style

Access Paper or Ask Questions