Search papers, labs, and topics across Lattice.
This paper introduces the Neuromodulated Constrained Autoencoder (NcAE), an extension of constrained autoencoders (cAEs) that incorporates a neuromodulatory mechanism to adapt geometric constraints based on contextual information. The NcAE uses gain and bias tuning, conditioned on static contextual information, to parameterize these geometric constraints. Experiments on dynamical systems demonstrate that NcAEs can accurately capture variations in manifold geometry across different regimes while maintaining projection properties, effectively decoupling global contextual parameters from local manifold representations.
Neuromodulation offers a way to disentangle global contextual parameters from local manifold representations in constrained autoencoders, enabling context-aware dimensionality reduction.
Constrained autoencoders (cAE) provide a successful path towards interpretable dimensionality reduction by enforcing geometric structure on latent spaces. However, standard cAEs cannot adapt to varying physical parameters or environmental conditions without conflating these contextual shifts with the primary input. To address this, we integrated a neuromodulatory mechanism into the cAE framework to allow for context-dependent manifold learning. This paper introduces the Neuromodulated Constrained Autoencoder (NcAE), which adaptively parameterizes geometric constraints via gain and bias tuning conditioned on static contextual information. Experimental results on dynamical systems show that the NcAE accurately captures how manifold geometry varies across different regimes while maintaining rigorous projection properties. These results demonstrate that neuromodulation effectively decouples global contextual parameters from local manifold representations. This architecture provides a foundation for developing more flexible, physics-informed representations in systems subject to (non-stationary) environmental constraints.
