stdp-bernoulli-message-passing

What problem does it solve?

STDP-driven Bernoulli messaging enables probabilistic inference on spike-based data by combining spike-timing-dependent plasticity with Bernoulli variable messaging in a factor-graph framework. This approach supports Bayesian reasoning on neuromorphic signals and compact probabilistic representations.

Core Features & Use Cases

• STDP-based weight updates enable biologically plausible learning within a Bernoulli message-passing network.
• Factor-graph-inspired inference supports decoding and probabilistic reasoning over small networks with uncertain observations.
• Use Case: simulate neuromorphic decoding or Bayesian inference tasks on toy data to study convergence and robustness.

Quick Start

Configure a small BernoulliMessageConfig, build a BernoulliMessagePassingNetwork, and run a few iterations to observe beliefs.