---
_id: '67308'
abstract:
- lang: eng
text: Optimizing large-scale multibody systems is a challenging task, particularly
in the presence of multiple conflicting criteria. To prevent high simulation costs,
surrogate models constructed from a small number of expensive model evaluations
are very popular. However, it is difficult to ensure the optimality of the obtained
solutions using a single pre-computed model. We present a back-and-forth approach
between surrogate modeling and multi-objective optimization, and we compare different
strategies for optimization, sampling, and surrogate modeling, to identify the
most promising approach in terms of computational efficiency and solution quality.
author:
- first_name: Augustina Chidinma
full_name: Amakor, Augustina Chidinma
id: '97916'
last_name: Amakor
- first_name: Manuel B.
full_name: Berkemeier, Manuel B.
last_name: Berkemeier
- first_name: Meike Claudia
full_name: Wohlleben, Meike Claudia
id: '43991'
last_name: Wohlleben
orcid: 0009-0009-9767-7168
- first_name: Walter
full_name: Sextro, Walter
id: '21220'
last_name: Sextro
- first_name: Sebastian
full_name: Peitz, Sebastian
id: '47427'
last_name: Peitz
orcid: 0000-0002-3389-793X
citation:
ama: 'Amakor AC, Berkemeier MB, Wohlleben MC, Sextro W, Peitz S. Surrogate-Assisted
Multi-objective Design of Complex Multibody Systems. In: Senn W, Sanguineti M,
Saudargiene A, et al., eds. Artificial Neural Networks and Machine Learning
– ICANN 2025. Springer Nature Switzerland; 2026:251–262. doi:10.1007/978-3-032-04555-3_21'
apa: Amakor, A. C., Berkemeier, M. B., Wohlleben, M. C., Sextro, W., & Peitz,
S. (2026). Surrogate-Assisted Multi-objective Design of Complex Multibody Systems.
In W. Senn, M. Sanguineti, A. Saudargiene, I. V. Tetko, A. E. P. Villa, V. Jirsa,
& Y. Bengio (Eds.), Artificial Neural Networks and Machine Learning – ICANN
2025 (pp. 251–262). Springer Nature Switzerland. https://doi.org/10.1007/978-3-032-04555-3_21
bibtex: '@inproceedings{Amakor_Berkemeier_Wohlleben_Sextro_Peitz_2026, place={Cham},
title={Surrogate-Assisted Multi-objective Design of Complex Multibody Systems},
DOI={10.1007/978-3-032-04555-3_21},
booktitle={Artificial Neural Networks and Machine Learning – ICANN 2025}, publisher={Springer
Nature Switzerland}, author={Amakor, Augustina Chidinma and Berkemeier, Manuel
B. and Wohlleben, Meike Claudia and Sextro, Walter and Peitz, Sebastian}, editor={Senn,
Walter and Sanguineti, Marcello and Saudargiene, Ausra and Tetko, Igor V. and
Villa, Alessandro E. P. and Jirsa, Viktor and Bengio, Yoshua}, year={2026}, pages={251–262}
}'
chicago: 'Amakor, Augustina Chidinma, Manuel B. Berkemeier, Meike Claudia Wohlleben,
Walter Sextro, and Sebastian Peitz. “Surrogate-Assisted Multi-Objective Design
of Complex Multibody Systems.” In Artificial Neural Networks and Machine Learning
– ICANN 2025, edited by Walter Senn, Marcello Sanguineti, Ausra Saudargiene,
Igor V. Tetko, Alessandro E. P. Villa, Viktor Jirsa, and Yoshua Bengio, 251–262.
Cham: Springer Nature Switzerland, 2026. https://doi.org/10.1007/978-3-032-04555-3_21.'
ieee: 'A. C. Amakor, M. B. Berkemeier, M. C. Wohlleben, W. Sextro, and S. Peitz,
“Surrogate-Assisted Multi-objective Design of Complex Multibody Systems,” in Artificial
Neural Networks and Machine Learning – ICANN 2025, 2026, pp. 251–262, doi:
10.1007/978-3-032-04555-3_21.'
mla: Amakor, Augustina Chidinma, et al. “Surrogate-Assisted Multi-Objective Design
of Complex Multibody Systems.” Artificial Neural Networks and Machine Learning
– ICANN 2025, edited by Walter Senn et al., Springer Nature Switzerland, 2026,
pp. 251–262, doi:10.1007/978-3-032-04555-3_21.
short: 'A.C. Amakor, M.B. Berkemeier, M.C. Wohlleben, W. Sextro, S. Peitz, in: W.
Senn, M. Sanguineti, A. Saudargiene, I.V. Tetko, A.E.P. Villa, V. Jirsa, Y. Bengio
(Eds.), Artificial Neural Networks and Machine Learning – ICANN 2025, Springer
Nature Switzerland, Cham, 2026, pp. 251–262.'
date_created: 2026-10-01T12:02:40Z
date_updated: 2026-10-01T12:03:16Z
department:
- _id: '655'
doi: 10.1007/978-3-032-04555-3_21
editor:
- first_name: Walter
full_name: Senn, Walter
last_name: Senn
- first_name: Marcello
full_name: Sanguineti, Marcello
last_name: Sanguineti
- first_name: Ausra
full_name: Saudargiene, Ausra
last_name: Saudargiene
- first_name: Igor V.
full_name: Tetko, Igor V.
last_name: Tetko
- first_name: Alessandro E. P.
full_name: Villa, Alessandro E. P.
last_name: Villa
- first_name: Viktor
full_name: Jirsa, Viktor
last_name: Jirsa
- first_name: Yoshua
full_name: Bengio, Yoshua
last_name: Bengio
keyword:
- own
- own-conference
language:
- iso: eng
page: 251–262
place: Cham
publication: Artificial Neural Networks and Machine Learning – ICANN 2025
publication_identifier:
isbn:
- 978-3-032-04555-3
publisher: Springer Nature Switzerland
status: public
title: Surrogate-Assisted Multi-objective Design of Complex Multibody Systems
type: conference
user_id: '47427'
year: '2026'
...
---
_id: '67303'
author:
- first_name: Augustina Chidinma
full_name: Amakor, Augustina Chidinma
id: '97916'
last_name: Amakor
- first_name: Konstantin
full_name: Sonntag, Konstantin
id: '56399'
last_name: Sonntag
orcid: https://orcid.org/0000-0003-3384-3496
- first_name: Sebastian
full_name: Peitz, Sebastian
id: '47427'
last_name: Peitz
orcid: 0000-0002-3389-793X
citation:
ama: 'Amakor AC, Sonntag K, Peitz S. Interactive Pareto navigation for deep multi-task
learning. In: European Conference on Machine Learning and Principles and Practice
of Knowledge Discovery in Databases (ECML PKDD). ; 2026:652-668. doi:10.1007/978-3-032-37667-1_37'
apa: Amakor, A. C., Sonntag, K., & Peitz, S. (2026). Interactive Pareto navigation
for deep multi-task learning. European Conference on Machine Learning and Principles
and Practice of Knowledge Discovery in Databases (ECML PKDD), 652–668. https://doi.org/10.1007/978-3-032-37667-1_37
bibtex: '@inproceedings{Amakor_Sonntag_Peitz_2026, title={Interactive Pareto navigation
for deep multi-task learning}, DOI={10.1007/978-3-032-37667-1_37},
booktitle={European Conference on Machine Learning and Principles and Practice
of Knowledge Discovery in Databases (ECML PKDD)}, author={Amakor, Augustina Chidinma
and Sonntag, Konstantin and Peitz, Sebastian}, year={2026}, pages={652–668} }'
chicago: Amakor, Augustina Chidinma, Konstantin Sonntag, and Sebastian Peitz. “Interactive
Pareto Navigation for Deep Multi-Task Learning.” In European Conference on
Machine Learning and Principles and Practice of Knowledge Discovery in Databases
(ECML PKDD), 652–68, 2026. https://doi.org/10.1007/978-3-032-37667-1_37.
ieee: 'A. C. Amakor, K. Sonntag, and S. Peitz, “Interactive Pareto navigation for
deep multi-task learning,” in European Conference on Machine Learning and Principles
and Practice of Knowledge Discovery in Databases (ECML PKDD), 2026, pp. 652–668,
doi: 10.1007/978-3-032-37667-1_37.'
mla: Amakor, Augustina Chidinma, et al. “Interactive Pareto Navigation for Deep
Multi-Task Learning.” European Conference on Machine Learning and Principles
and Practice of Knowledge Discovery in Databases (ECML PKDD), 2026, pp. 652–68,
doi:10.1007/978-3-032-37667-1_37.
short: 'A.C. Amakor, K. Sonntag, S. Peitz, in: European Conference on Machine Learning
and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2026,
pp. 652–668.'
date_created: 2026-10-01T11:56:15Z
date_updated: 2026-10-01T11:58:43Z
department:
- _id: '655'
doi: 10.1007/978-3-032-37667-1_37
keyword:
- own
- own-conference
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2606.19521
oa: '1'
page: 652-668
publication: European Conference on Machine Learning and Principles and Practice of
Knowledge Discovery in Databases (ECML PKDD)
status: public
title: Interactive Pareto navigation for deep multi-task learning
type: conference
user_id: '47427'
year: '2026'
...
---
_id: '51159'
abstract:
- lang: eng
text: Sparsity is a highly desired feature in deep neural networks (DNNs) since
it ensures numerical efficiency, improves the interpretability of models (due
to the smaller number of relevant features), and robustness. In machine learning
approaches based on linear models, it is well known that there exists a connecting
path between the sparsest solution in terms of the $\ell^1$ norm,i.e., zero weights
and the non-regularized solution, which is called the regularization path. Very
recently, there was a first attempt to extend the concept of regularization paths
to DNNs by means of treating the empirical loss and sparsity ($\ell^1$ norm) as
two conflicting criteria and solving the resulting multiobjective optimization
problem. However, due to the non-smoothness of the $\ell^1$ norm and the high
number of parameters, this approach is not very efficient from a computational
perspective. To overcome this limitation, we present an algorithm that allows
for the approximation of the entire Pareto front for the above-mentioned objectives
in a very efficient manner. We present numerical examples using both deterministic
and stochastic gradients. We furthermore demonstrate that knowledge of the regularization
path allows for a well-generalizing network parametrization.
article_number: '100625'
author:
- first_name: Augustina Chidinma
full_name: Amakor, Augustina Chidinma
id: '97916'
last_name: Amakor
- first_name: Konstantin
full_name: Sonntag, Konstantin
id: '56399'
last_name: Sonntag
orcid: https://orcid.org/0000-0003-3384-3496
- first_name: Sebastian
full_name: Peitz, Sebastian
id: '47427'
last_name: Peitz
orcid: 0000-0002-3389-793X
citation:
ama: Amakor AC, Sonntag K, Peitz S. A multiobjective continuation method to compute
the regularization path of deep neural networks. Machine Learning with Applications.
2025;19. doi:10.1016/j.mlwa.2025.100625
apa: Amakor, A. C., Sonntag, K., & Peitz, S. (2025). A multiobjective continuation
method to compute the regularization path of deep neural networks. Machine
Learning with Applications, 19, Article 100625. https://doi.org/10.1016/j.mlwa.2025.100625
bibtex: '@article{Amakor_Sonntag_Peitz_2025, title={A multiobjective continuation
method to compute the regularization path of deep neural networks}, volume={19},
DOI={10.1016/j.mlwa.2025.100625},
number={100625}, journal={Machine Learning with Applications}, publisher={Elsevier},
author={Amakor, Augustina Chidinma and Sonntag, Konstantin and Peitz, Sebastian},
year={2025} }'
chicago: Amakor, Augustina Chidinma, Konstantin Sonntag, and Sebastian Peitz. “A
Multiobjective Continuation Method to Compute the Regularization Path of Deep
Neural Networks.” Machine Learning with Applications 19 (2025). https://doi.org/10.1016/j.mlwa.2025.100625.
ieee: 'A. C. Amakor, K. Sonntag, and S. Peitz, “A multiobjective continuation method
to compute the regularization path of deep neural networks,” Machine Learning
with Applications, vol. 19, Art. no. 100625, 2025, doi: 10.1016/j.mlwa.2025.100625.'
mla: Amakor, Augustina Chidinma, et al. “A Multiobjective Continuation Method to
Compute the Regularization Path of Deep Neural Networks.” Machine Learning
with Applications, vol. 19, 100625, Elsevier, 2025, doi:10.1016/j.mlwa.2025.100625.
short: A.C. Amakor, K. Sonntag, S. Peitz, Machine Learning with Applications 19
(2025).
date_created: 2024-02-06T08:51:00Z
date_updated: 2026-10-01T11:33:45Z
ddc:
- '000'
department:
- _id: '655'
doi: 10.1016/j.mlwa.2025.100625
file:
- access_level: closed
content_type: application/pdf
creator: speitz
date_created: 2026-10-01T11:24:57Z
date_updated: 2026-10-01T11:24:57Z
file_id: '67296'
file_name: ASP25_A multiobjective continuation method to compute the regularization
path of deep neural networks.pdf
file_size: 2433955
relation: main_file
success: 1
file_date_updated: 2026-10-01T11:24:57Z
has_accepted_license: '1'
intvolume: ' 19'
language:
- iso: eng
publication: Machine Learning with Applications
publisher: Elsevier
quality_controlled: '1'
status: public
title: A multiobjective continuation method to compute the regularization path of
deep neural networks
type: journal_article
user_id: '47427'
volume: 19
year: '2025'
...