← Authors

Philipp Becker

14 papers · 2019–2025 · 6 conferences · across top CS/AI conferences

Achievements

🐝 Cross-Pollinator (15) πŸ—ΊοΈ Taxonomy Completionist (16) 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird 🌍 Conference Polyglot (6) πŸƒ Academic Marathon (6) πŸŒ‰ Interdisciplinary Bridge 🀝 Dynamic Duo (13) πŸ’Ž Century Club (14) πŸ”₯ Unstoppable (7)

Conferences

ICLR (6) CORL (4) ICCV (1) ICML (1) NIPS (1) RSS (1)

Top co-authors

Gerhard Neumann (13) Vaisakh Shaj (3) Vien Anh Ngo (3) Michael Volpp (2) Niels van Duijkeren (2) C. James Taylor (2) Dieter BΓΌchler (2) Harit Pandya (2) Fabian Otto (2) Niklas Freymuth (1)

Keywords

recurrent kalman network (2) inverse dynamics (2) image generation (1) imitation learning (1) policy optimization (1) maximum entropy (1) variational inference (1) text-to-image synthesis (1) distribution matching (1) kalman filter (1) distribution shift (1) trajectory-based planning (1) probabilistic model (1) recurrent neural network (1) generalization capability (1) mixture of expert (1) uncertainty estimation (1) bayesian deep learning (1) linear attention (1) diffusion transformer (1)

Papers

Efficient Off-Policy Learning for High-Dimensional Action Spaces ICLR 2025
Geometry-aware RL for Manipulation of Varying Shapes and Deformable Objects ICLR 2025
EDiT: Efficient Diffusion Transformers with Linear Compressed Attention ICCV 2025
PointPatchRL - Masked Reconstruction Improves Reinforcement Learning on Point Clouds CORL 2024
Accurate Bayesian Meta-Learning by Accurate Task Posterior Inference ICLR 2023
Beyond Deep Ensembles: A Large-Scale Evaluation of Bayesian Deep Learning under Distribution Shift NIPS 2023
End-to-End Learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control RSS 2022
Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors CORL 2022
Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarios ICLR 2022
Differentiable Trust Region Layers for Deep Reinforcement Learning ICLR 2021
Specializing Versatile Skill Libraries using Local Mixture of Experts CORL 2021
Expected Information Maximization: Using the I-Projection for Mixture Density Estimation ICLR 2020
Action-Conditional Recurrent Kalman Networks For Forward and Inverse Dynamics Learning CORL 2020
Recurrent Kalman Networks: Factorized Inference in High-Dimensional Deep Feature Spaces ICML 2019