Search papers, labs, and topics across Lattice.
This paper introduces MetaLearnNCA, a novel few-shot offline meta-learning framework that leverages interacting Neural Cellular Automata (NCAs) to perform task adaptation without explicit gradient computation during inference. The system comprises an Active-NCA for task inference and a Meta-NCA that dynamically updates a 2D spatial program by diffusing error residuals. MetaLearnNCA demonstrates competitive in-distribution performance and significant out-of-distribution transfer gains, suggesting the potential of decentralized cellular dynamics for gradient-free meta-learning.
Gradient-free few-shot adaptation emerges from decentralized cellular dynamics, achieving state-of-the-art OOD transfer by updating a spatial program through local neighborhood interactions.
Few-shot meta-learning traditionally formulates task adaptation either as analytical gradient descent through unrolled computational graphs or as metric-based distance comparisons over flattened 1D fea- ture vectors, which either incur costly test-time backpropagation or discard native 2D spatial geometry. In this work, we propose METALEARNNCA, a decentralized framework that achieves few-shot adapta- tion through the dynamical interaction of coupled Neural Cellular Automata (NCAs) without computing analytical gradients during inference. MetaLearnNCA decomposes task adaptation into an Active- NCA, which executes task inference conditioned on a continuous 2D spatial memory grid termed the spatial program, and a learned Meta-NCA, which acts as a decentralized cellular optimizer by diffusing spatial error residuals across local neighborhoods to dynamically update this program. METALEARN- NCA is competitive against canonical meta-learners in-distribution (96.12% on Omniglot) with Out-Of- Distribution transfer gains on MNIST, KMNIST, and Fashion-MNIST transfer across 10 independent testing seeds across 1-, 5-, and 10-shot regimes (e.g., surpassing Prototypical Networks by +10.54% on 10-shot MNIST and a +3.87% gain on 10-shot Fashion-MNIST over FOMAML). Our results establish that robust, gradient-free learning-to-learn can emerge from decentralized cellular dynamics on non-von Neumann substrates.
