Article

Learning sculpts orthogonal task manifolds for continual skill learning in recurrent networks

2026-02-16

Abstract excerpt

Humans and animals can learn and seamlessly perform a vast repertoire of behaviors. However, how neural populations incorporate new skills without disrupting previously learned ones remains poorly understood. This challenge is known as catastrophic forgetting in artificial neural networks and is especially severe in recurrent networks, where computation relies on stable internal dynamics. While current machine lea...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: d05fa01e-b11e-58a3-acd0-25867f4f17d8
DOI: 10.64898/2026.02.15.705283

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Learning sculpts orthogonal task manifolds for continual skill learning in recurrent networksDOI 10.64898/2026.02.15.705283Read

Flexible multitask computation in recurrent networks utilizes shared dynamical motifsSemantic proximity Read

A neural mechanism for online discovery of latent contextsSemantic proximity Read

Scaffolded representation learning in deep networksSemantic proximity Read

Nonlinear manifolds underlie neural population activity during behaviourSemantic proximity Read

Overcoming Catastrophic Interference: Neuroscience-inspired Models for Continuous Learning in Neural NetworksSemantic proximity Read

Neuro-inspired cascaded memory coordination enables continual generalizationSemantic proximity Read

Multitasking Recurrent Networks Utilize Compositional Strategies for Control of MovementSemantic proximity Read

Adaptive proximity to criticality underlies amplification of ultra-slow fluctuations during free recallSemantic proximity Read

Memory Consolidation with Orthogonal Gradients for avoiding Catastrophic ForgettingSemantic proximity Read

Compositional Recombination Relies on a Distributed Cortico-Cerebellar NetworkSemantic proximity Read

Abstract representations emerge naturally in neural networks trained to perform multiple tasksSemantic proximity Read

Compositional Recombination Relies on a Distributed Cortico-Cerebellar NetworkSemantic proximity Read