Sources

[1] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). "Learning Internal Representations by Error Propagation." Nature. — The original backprop paper that launched the field.

[2] Lillicrap, T. P., et al. (2016). "Random synaptic feedback weights support error backpropagation for deep learning." Nature Communications. — Discovered feedback alignment, removing the weight symmetry requirement as a fatal objection to biologically plausible backprop.

[3] Scellier, B., & Bengio, Y. (2017). "Equilibrium Propagation: Bridging the Gap Between Energy-Based Models and Backpropagation." Frontiers in Computational Neuroscience. — The foundational EqProp paper showing that energy-based models can compute gradients using local Hebbian-like updates during inference.

[4] Kendall, J. (2026). "Equilibrium Propagation and Hamiltonian Inference in the Diffusive FitzHugh-Nagumo Model." arXiv:2605.21568. Zyphra. — Extended EqProp from energy-based models to biologically realistic neurons (FitzHugh-Nagumo), showing learning through physical dynamics.

[5] Hinton, G. E. (2022). "The Forward-Forward Algorithm: Some Preliminary Investigations." arXiv:2212.13345. — Hinton's alternative to backprop using two forward passes with a local "goodness" objective, eliminating the backward pass entirely.

[6] Rao, R. P. N., & Ballard, D. H. (1999). "Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects." Nature Neuroscience. — The original neuroscience paper proposing predictive coding as a model of visual cortex function.

[7] Whittington, J. C. R., & Bogacz, R. (2017). "An Approximation of the Error Backpropagation Algorithm in a Predictive Coding Network with Local Hebbian Synaptic Plasticity." Neural Computation. — Showed that predictive coding approximates backprop with local learning rules.

[8] Salvatori, T., et al. (2023/2025). "Brain-inspired Computational Intelligence via Predictive Coding." arXiv:2308.07870. — Comprehensive predictive coding survey covering history, results, and implications. Authors include Friston, Rao, and key ML researchers.

[9] Lv, C., et al. (2024). "Towards Biologically Plausible Computing: A Comprehensive Comparison." arXiv:2406.16062. — The most comprehensive survey of biologically plausible algorithms, comparing 9 candidate algorithms against actual brain activity.

[10] Bengio, Y., et al. (2015). "Towards Biologically Plausible Deep Learning." arXiv:1502.04156. — Early manifesto identifying the three implausibilities of backprop and proposing STDP-based alternatives within a variational EM framework.

[11] Wolters, C., et al. (2022). "Biologically Plausible Learning on Neuromorphic Hardware Architectures." arXiv:2212.14337. — Hardware co-design perspective comparing backprop vs feedback alignment on compute-in-memory architectures.

[12] Human Unsupervised (2025). "Neuromorphic Computing 2025: Current State of the Art." humanunsupervised.com. — Up-to-date survey of neuromorphic hardware progress: Loihi 2, memristors, spintronics, photonics.

[13] Zyphra (2026). "Equilibrium Propagation Beyond Energy-Based Models." zyphra.com/blog. — Non-technical summary of the FitzHugh-Nagumo EqProp result, explaining the implications for AI hardware and biologically plausible learning.