# Joint Surrogate Learning of Objectives, Constraints, and Sensitivities for Efficient Multi-objective Optimization of Neural Dynamical Systems

Source: [Joint Surrogate Learning of Objectives, Constraints, and Sensitivities for Efficient Multi-objective Optimization of Neural Dynamical Systems](<https://www.sota2.com/research/paper/69c21a750636e5bf4afc477c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-25T05:00:04.214Z

Authors: Frithjof Gressmann, Ivan Georgiev Raikov, Seung Hyun Kim, Mattia Gazzola, Lawrence Rauchwerger, Ivan Soltesz

Published: 2026-03-22T00:06:28.000Z

## Abstract

Biophysical neural system simulations are among the most computationally demanding scientific applications, and their optimization requires navigating high-dimensional parameter spaces under numerous constraints that impose a binary feasible/infeasible partition with no gradient signal to guide the search. Here, we introduce DMOSOPT, a scalable optimization framework that leverages a unified, jointly learned surrogate model to capture the interplay between objectives, constraints, and parameter sensitivities. By learning a smooth approximation of both the objective landscape and the feasibility boundary, the joint surrogate provides a unified gradient that simultaneously steers the search toward improved objective values and greater constraint satisfaction, while its partial derivatives yield per-parameter sensitivity estimates that enable more targeted exploration. We validate the framework from single-cell dynamics to population-level network activity, spanning incremental stages of a neural circuit modeling workflow, and demonstrate efficient, effective optimization of highly constrained problems at supercomputing scale with substantially fewer problem evaluations. While motivated by and demonstrated in the context of computational neuroscience, the framework is general and applicable to constrained multi-objective optimization problems across scientific and engineering domains.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Surrogate-assisted optimization on CA1 NGFC (Neurogliaform cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-ngfc-neurogliaform-cell.md>): Surrogate-assisted optimization; CA1 NGFC (Neurogliaform cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 BS (Bistratified cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-bs-bistratified-cell.md>): Surrogate-assisted optimization; CA1 BS (Bistratified cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 SCA (Schaffer collateral-associated cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-sca-schaffer-collateral-associated-cell.md>): Surrogate-assisted optimization; CA1 SCA (Schaffer collateral-associated cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 OLM (Oriens-lacunosum moleculare cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-olm-oriens-lacunosum-moleculare-cell.md>): Surrogate-assisted optimization; CA1 OLM (Oriens-lacunosum moleculare cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 PVBC (Parvalbumin-positive basket cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-pvbc-parvalbumin-positive-basket-cell.md>): Surrogate-assisted optimization; CA1 PVBC (Parvalbumin-positive basket cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 AAC (Axo-axonic cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-aac-axo-axonic-cell.md>): Surrogate-assisted optimization; CA1 AAC (Axo-axonic cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 CCKBC (Cholecystokinin-positive basket cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-cckbc-cholecystokinin-positive-basket-cell.md>): Surrogate-assisted optimization; CA1 CCKBC (Cholecystokinin-positive basket cell); Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 IVY cell](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-ivy-cell.md>): Surrogate-assisted optimization; CA1 IVY cell; Normalized Hypervolume: not recorded.

- [Surrogate-assisted optimization on CA1 IS (Interneuron-specific cell)](<https://www.sota2.com/research/sota/surrogate-assisted-optimization-on-ca1-is-interneuron-specific-cell.md>): Surrogate-assisted optimization; CA1 IS (Interneuron-specific cell); Normalized Hypervolume: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.20984>)

[PDF](<https://arxiv.org/pdf/2603.20984.pdf>)
