Predictive coding equals proximal gradient descent on MAP problems, with priors setting nonlinearities via proximal operators and yielding leaky firing-rate networks plus hierarchical MRFs.
arXiv preprint arXiv:2107.12979 , year=
10 Pith papers cite this work, alongside 26 external citations. Polarity classification is still indexing.
10
Pith papers citing it
26
external citations · external index
verdicts
UNVERDICTED 10representative citing papers
A VGG10 predictive coding network is trained on ImageNet via equilibrium propagation to 13.23% top-5 error, close to the 12.2% backpropagation baseline, marking the first such demonstration at this scale.
PC-ALM uses dual ascent on an augmented Lagrangian to achieve exact backpropagation gradients via layer-local updates in linear networks and matching performance in nonlinear networks up to depth 128.
Latent prediction SSL recovers latent trees from PCFG data with sample complexity constant in hierarchy depth L (up to logs), unlike exponential for token-level or supervised methods.
Extends equilibrium propagation to skew-gradient Fitzhugh-Nagumo systems and derives an explicit layer-wise Hamiltonian recurrence for inference in deep residual topologies.
Retrieval from out-of-domain foundation models enables personalization of a lightweight transformer for stress detection, yielding +3.92% accuracy and +4.76% F1 gains on WESAD without user labels.
Predictive coding is recast as deep hierarchical Gaussian filters to restore precision-weighted message passing, yielding closed-form inference and online precision learning that matches backpropagation speed on FashionMNIST while outperforming on online and concept-drift tasks.
SAL is a spike-timing-based local learning rule that aligns feedback weights to forward weights in spiking networks by exploiting noise and Hebbian/anti-Hebbian plasticity to recover the true gradient.
A revised DMBN with positional time encoding improves temporal representation and generalization in neural processes for multimodal robotic action prediction.
Promise Theory is augmented with Bayesian methods and information optimization to model intentionality, inference, and swarm formation in autonomous agents.
citing papers explorer
-
Predictive Coding with Bayesian Priors via Proximal Gradients
Predictive coding equals proximal gradient descent on MAP problems, with priors setting nonlinearities via proximal operators and yielding leaky firing-rate networks plus hierarchical MRFs.
-
Training a Predictive Coding Network on ImageNet using Equilibrium Propagation
A VGG10 predictive coding network is trained on ImageNet via equilibrium propagation to 13.23% top-5 error, close to the 12.2% backpropagation baseline, marking the first such demonstration at this scale.
-
Augmented Lagrangian Predictive Coding
PC-ALM uses dual ascent on an augmented Lagrangian to achieve exact backpropagation gradients via layer-local updates in linear networks and matching performance in nonlinear networks up to depth 128.
-
Learn from your own latents and not from tokens: A sample-complexity theory
Latent prediction SSL recovers latent trees from PCFG data with sample complexity constant in hierarchy depth L (up to logs), unlike exponential for token-level or supervised methods.
-
Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Model
Extends equilibrium propagation to skew-gradient Fitzhugh-Nagumo systems and derives an explicit layer-wise Hamiltonian recurrence for inference in deep residual topologies.
-
Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detection
Retrieval from out-of-domain foundation models enables personalization of a lightweight transformer for stress detection, yielding +3.92% accuracy and +4.76% F1 gains on WESAD without user labels.
-
Closed-form predictive coding via hierarchical Gaussian filters
Predictive coding is recast as deep hierarchical Gaussian filters to restore precision-weighted message passing, yielding closed-form inference and online precision learning that matches backpropagation speed on FashionMNIST while outperforming on online and concept-drift tasks.
-
Spike-based alignment learning solves the weight transport problem
SAL is a spike-timing-based local learning rule that aligns feedback weights to forward weights in spiking networks by exploiting noise and Hebbian/anti-Hebbian plasticity to recover the true gradient.
-
Exploring Temporal Representation in Neural Processes for Multimodal Action Prediction
A revised DMBN with positional time encoding improves temporal representation and generalization in neural processes for multimodal robotic action prediction.
-
Quantitative Promise Theory: Intentionality and Inference in Autonomous Agents
Promise Theory is augmented with Bayesian methods and information optimization to model intentionality, inference, and swarm formation in autonomous agents.