M3 Optimise
State- and Parameter-Space Exploration and Process Optimisation
M3 finds good process settings with as few trials as possible, using what can be observed between process stages.
Work packages
Results of the first funding phase, by work package.
-
M3.1 Simulated benchmark environment for multi-stage exploration and optimisation
A simulation package for multi-stage systems defines subprocesses with their inputs, outputs and noise, and connects them into a directed acyclic graph. It works with scalable Bayesian optimisation packages such as BoTorch and with GPU-accelerated inference.
Enlarge figure A multi-stage process as a chain of subprocesses, each with its own inputs and observable outputs. Figure: KI-FOR 5339. Publications
-
Bayesian Optimization using Partially Observable Gaussian Process Network
NeurIPS Workshop on ML×OR, 2025
-
Virtual Process Dossier: A Process-Aware Data Catalogue
arXiv preprint, 2026
-
-
M3.2 Multi-stage Bayesian optimisation
The Partially Observable Gaussian Process Network (POGPN) treats the output of each node as latent and only partially observed, so it propagates uncertainty correctly across the whole process network and models noisy intermediate observations. It is trained with variational inference and runs with BoTorch.
Publications
-
Bayesian Optimization using Partially Observable Gaussian Process Network
NeurIPS Workshop on ML×OR, 2025
-
-
M3.3 Open-loop trajectory exploration and optimisation
Data from a non-isothermal glass moulding process served as a real multi-stage test case, evaluated with a new validation strategy that combines stratified sampling and bootstrapping. Joint parameter and state-space (JPSS) modelling extracts latent features from intermediate time series and consistently outperformed direct input-output models. The JPSS paper received the Best Student Paper Award in the AI for Industry track at IEEE INDIN 2024.
-
M3.4 Robust multi-stage exploration and optimisation
A trust-region method based on Fisher information works with any model that has a smooth, differentiable posterior. It comes with a theoretical analysis of vanishing gradients in high-dimensional Bayesian optimisation (NeurIPS 2026).
Publications
-
M3.5 Joint state-space and parameter-space exploration and optimisation
POGPN outperformed conventional Gaussian process networks and Bayesian optimisation on high-dimensional benchmark functions and a penicillin production simulation, with both noise-free and noisy observations. Combined with JPSS, it was tested on high-dimensional time series of a multi-stage bioethanol process.
-
M3.6 Application to the physical composite stamp-forming process
In a high-fidelity Abaqus forming simulation, Bayesian optimisation of gripper positions and forces reduced out-of-plane wrinkling of carbon-fibre thermoplastic laminates, and trust-region methods placed an additional gripper. A sensitivity analysis showed that gripper positions matter more than gripper forces.
Publications
-
Composites Part A: Applied Science and Manufacturing, 2026
-
A result, jointly with M1, T1 and T2
Random gripper layout
Layout found by Bayesian optimisation
About the subproject
Every real trial is expensive, so each one should be as informative as possible. Bayesian optimisation proposes the next trial by weighing what the model already knows against where it is still uncertain.
- Principal investigators
- Researchers
-
- Saksham Kiroriwal
- Host institute
- Fraunhofer IOSB
Publications 8
-
Bayesian Optimization with Fisher Information Geometry: Gradient Bounds and Trust-Region Methods
-
BOING: Optimizing Bayesian Optimization with Information Geometry
-
Bayesian Optimization using Partially Observable Gaussian Process Network
-
Joint parameter and state-space modelling of manufacturing processes using Gaussian processes