Research Paper#Deep Learning, Recurrent Neural Networks, Biological Plausibility, Online Learning🔬 ResearchAnalyzed: Jan 3, 2026 09:24
Extending E-prop for Deep Recurrent Networks
Published:Dec 30, 2025 23:10
•1 min read
•ArXiv
Analysis
This paper addresses the biological implausibility of Backpropagation Through Time (BPTT) in training recurrent neural networks. It extends the E-prop algorithm, which offers a more biologically plausible alternative to BPTT, to handle deep networks. This is significant because it allows for online learning of deep recurrent networks, mimicking the hierarchical and temporal dynamics of the brain, without the need for backward passes.
Key Takeaways
- •Extends E-prop, a biologically plausible alternative to BPTT, to deep recurrent networks.
- •Enables online learning of deep recurrent networks.
- •Avoids the need for backpropagation through time.
- •Addresses the limitations of BPTT in terms of biological plausibility.
Reference
“The paper derives a novel recursion relationship across depth which extends the eligibility traces of E-prop to deeper layers.”