AutoHebb networks mimic human learning with self-modifying neural architecture · 2 sources tracked

Researchers have introduced AutoHebb, a novel class of self-modifying neural networks designed to mimic human and animal learning capabilities like zero-shot inference and continual learning. Unlike traditional models that rely on gradient descent, AutoHebb networks integrate external feedback with internally generated signals to create a global learning modulator. This modulator then governs a biologically grounded 3-factor Hebbian rule for updating network weights, enabling both simplicity and expressiveness. Experiments show AutoHebb networks excel at zero-shot inference and continual learning tasks, outperforming current models and demonstrating emergent learning abilities. AI

IMPACT Introduces a novel biologically-inspired learning mechanism that could advance zero-shot and continual learning capabilities in AI systems.

RANK_REASON The cluster describes a new research paper detailing a novel neural network architecture and its performance on specific learning tasks.

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• AutoHebb
• CatalyzeX Code Finder for Papers
• Connected Papers
• DagsHub
• Gotit.pub
• gradient descent
• Hebbian rule
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• model release

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →