Research
Bahareh Tasdighi, Manuel Haussmann, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
European Conference on Artificial Intelligence (ECAI), 2025
We propose a novel actor-critic algorithm for continuous control in sparse reward settings, using a critic ensemble and PAC-Bayesian bounds on Bellman operator error to enable deep exploration via posterior sampling.
Gülçin Baykal, Abdullah Akgül, Manuel Haussmann, Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
Preprint on arXiv, 2025
Our library for RL research (get it here)
Bahareh Tasdighi, Manuel Haussmann, Yi-Shan Wu, Andres R. Masegosa, Melih Kandemir
Preprint on arXiv, 2025
Recursive PAC-Bayesian bounds for actor-critic methods.
Priscilla Ong, Manuel Haussmann, Otto Lönnroth, Harri Lähdesmäki
Transactions on Machine Learning Research (TMLR); published version, 2025
Latent mixed-effect models for high-dimensional longitudinal data
Machine Learning for Health Symposium; extended version, 2024
Learning high-dimensional mixed models via amortized variational inference
ICML Workshop on Structured Probabilistic Inference & Generative Modeling, 2024
A structured variational auto-encoder constrained with a mixed model for longitudinal data.
Abdullah Akgül, Gülçin Baykal, Manuel Haussmann, Melih Kandemir
Preprint on arXiv, 2025
Fixing plasticity issues in PPO via evidential learning.
Siddharth Ramchandran, Manuel Haussmann, Harri Lähdesmäki
International Conference on Learning Representations (ICLR), 2025
Improving Bayesian optimization by incorporating prior knowledge into the latent space.
Henning Bahl, Nina Elmer, Luigi Favaro, Manuel Haussmann, Tilman Plehn, Ramon Winterhalder
Preprint on arXiv, 2024
An evaluation of uncertainty calibration with BNNs and ensembles on LHC physics data.
Abdullah Akgül, Manuel Haussmann, Melih Kandemir
Conference on Neural Information Processing Systems (NeurIPS), 2024
We introduce a moment-matching based uncertainty penalty for the Bellman operator in an offline RL setting.
Bahareh Tasdighi, Abdullah Akgül, Manuel Haussmann, Kenny Kazimirzak Brink, Melih Kandemir
Advances in Approximate Bayesian Inference (AABI), 2024
We introduce the first PAC Bayesian bound trainable in a modern deep reinforcement learning pipeline to solve continuous control tasks.
Maksim Sinelnikov, Manuel Haussmann, Harri Lähdesmäki
International Conference on Machine Learning (ICML), 2024
A model for longitudinal data that can handle irregular observations and model missing not at random data.
Manuel Haussmann, Tran Minh Son Le, Viivi Halla-aho, Samu Kurki, Jussi Leinonen, Miika Koskinen, Samuel Kaski, Harri Lähdesmäki
International Conference on Artificial Intelligence and Statistics (AISTATS), 2024
We introduce a probabilistic model to combine different patient populations for trial effect estimation.
Samu Kurki*, Viivi Halla-Aho*, Manuel Haussmann, Harri Lähdesmäki, Jussi V Leinonen, Miika Koskinen
Nature Scientific Reports, 2024
An evaluation on the differences between a curated medical trial data set and uncurated electronic health records data.
Daolang Huang, Manuel Haussmann, Ulpu Remes, ST John, Grégoire Clarté, Kevin Luck, Samuel Kaski, Luigi Acerbi
Conference on Neural Information Processing Systems (NeurIPS), 2023
We introduce a method to incorporate high-dimensional equivariances into (conditional) neural processes.
Marco Bellagente, Manuel Haussmann, Michel Luchmann, Tilman Plehn
SciPost Physics, 2022
An evaluation of Bayesian normalizing flow for LHC events.
Melih Kandemir, Abdullah Akgül, Manuel Haussmann, Gözde Ünal
International Conference on Learning Representations (ICLR), 2022
Evidential deep learning combined with concepts from neural Turing machines and neural processes.
Manuel Haussmann*, Sebastian Gerwinn*, Andreas Look, Barbara Rakitsch, Melih Kandemir
International Conference on Artificial Intelligence and Statistics (AISTATS), 2021
We introduce a probabilistic method to learn dynamical systems that allows for partial input incorporation.
Manuel Haussmann, Sebastian Gerwinn, Melih Kandemir
Advances in Approximate Bayesian Inference (AABI), 2021
We combine evidential deep learning and PAC-based regularization with moment-matching approaches from Bayesian deep learning.
Sven Bollweg, Manuel Haussmann, Gregor Kasieczka, Michel Luchmann, Tilman Plehn, Jennifer Thompson
SciPost Physics, 2020
We discuss Bayesian deep-learning methods for jet classification in particle physics.
Manuel Haussmann, Fred A. Hamprecht, Melih Kandemir
Uncertainty in Artificial Intelligence Conference (UAI), 2020
A sampling-free approach to BNN learning via ReLU decomposition.
Manuel Haussmann, Fred A. Hamprecht, Melih Kandemir
International Joint Conference on Artificial Intelligence (IJCAI), 2019
We introduce a way to adaptively learn the acquisition function for active learning with reinforcement feedback.
Elke Kirschbaum, Manuel Haussmann, Steffen Wolf, Hannah Sonntag, Justus Schneider, Shehabeldin Elzoheiry, Oliver Kann, Daniel Durstewitz, Fred A Hamprecht
International Conference on Learning Representations (ICLR), 2019
We present LeMoNADe, an end-to-end learned motif detection method directly operating on calcium imaging videos.
Manuel Haussmann, Fred A. Hamprecht, Melih Kandemir
Conference on Computer Vision and Pattern Recognition (CVPR), 2017
Instance level inference with closed-form updates in a Gaussian processed-based model for weakly supervised multiple instance learning and object detection.
Melih Kandemir, Manuel Haussmann, Ferran Diego, Kumar Rajamani, Jeroen Van Der Laak, Fred Hamprecht
British Machine Vision Conference (BMVC), 2016
We introduce a model for multiple instance learning with Gaussian processes for bag-level predictions with closed-form updates.