Adaptive control of recurrent neural networks using conceptors
Pourcel, Guillaume; Goldmann, Mirko; Fischer, Ingo; Soriano, Miguel C.
Chaos 34, 103127 (2024)
Chaos 34, 103127 (2024)
Recurrent neural networks excel at predicting and generating complex high-dimensional temporal patterns. Due to their inherent nonlinear dynamics and memory, they can learn unbounded temporal dependencies from data. In a machine learning setting, the network’s parameters are adapted during a training phase to match the requirements of a given task/problem increasing its computational capabilities. After the training, the network parameters are kept fixed to exploit the learned computations. The static parameters, therefore, render the network unadaptive to changing conditions, such as an external or internal perturbation. In this paper, we demonstrate how keeping parts of the network adaptive even after the training enhances its functionality and robustness. Here, we utilize the conceptor framework and conceptualize an adaptive control loop analyzing the network’s behavior continuously and adjusting its time-varying internal representation to follow a desired target. We demonstrate how the added adaptivity of the network supports the computational functionality in three distinct tasks: interpolation of temporal patterns, stabilization against partial network degradation, and robustness against input distortion. Our results highlight the potential of adaptive networks in machine learning beyond training, enabling them to not only learn complex patterns but also dynamically adjust to changing environments, ultimately broadening their applicability.
Related research projects
NEHIL
Neuromorphic-Enhanced Heterogeneously-Integrated FMCW LiDAR
I.P.: Miguel C. Soriano
The NEHIL project, an EU-Korea partnership, is set to transform the landscape of digital technologies through groundbreaking neuromorphic architectures and advanced heterogeneous integration such as LiDAR systems. This collaborative initiative aims to …
POST-DIGITAL
POST-DIGITAL: Neuromorphic computing in photonic and other nonlinear media
I.P.: Ingo Fischer, Claudio Mirasso
POST-DIGITAL is a Marie Skłodowska-Curie Innovative Training Network, funded by the European Union’s Horizon 2020 research and innovation programme. POST-DIGITAL is committed to form a new generation of engineers and researchers, affording …
Noticias relacionadas
Mirko Goldmann rep un premi internacional per una recerca realitzada a l’IFISC
22 de enero de 2026
El guardó reconeix una recerca desenvolupada a l’IFISC que demostra que les xarxes neuronals recurrents poden continuar sent adaptatives més enllà de la fase d’entrenament.L’estudi premiat introdueix un bucle de control adaptatiu basat en conceptors que millora la robustesa i …
Mirko Goldmann receives international award for research carried out at IFISC
22 de enero de 2026
The award recognizes research carried out at IFISC showing that recurrent neural networks can remain adaptive beyond the training phase.The awarded study introduces a conceptor-based adaptive control loop that enhances network robustness and performance under perturbations.Researcher Mirko Goldmann has been …
Mirko Goldmann recibe un premio internacional por una investigación realizada en el IFISC
22 de enero de 2026
El galardón reconoce una investigación desarrollada en el IFISC que demuestra que las redes neuronales recurrentes pueden seguir siendo adaptativas más allá de la fase de entrenamiento.El estudio premiado introduce un bucle de control adaptativo basado en conceptors que mejora …