Deterministic versus stochastic dynamical classifiers: opposing random adversarial attacks with noise
This work addresses the problem of adversarial robustness in dynamical classifiers for neuroscience and machine learning applications, though it appears incremental as it builds on existing CVFR models.
The authors trained a Continuous-Variable Firing Rate (CVFR) model as a dynamical classifier by embedding attractors in its coupling matrix to sculpt basins of attraction for classification, and found that a stochastic variant of this model is robust to random adversarial attacks that corrupt input items.
The Continuous-Variable Firing Rate (CVFR) model, widely used in neuroscience to describe the intertangled dynamics of excitatory biological neurons, is here trained and tested as a veritable dynamically assisted classifier. To this end the model is supplied with a set of planted attractors which are self-consistently embedded in the inter-nodes coupling matrix, via its spectral decomposition. Learning to classify amounts to sculp the basin of attraction of the imposed equilibria, directing different items towards the corresponding destination target, which reflects the class of respective pertinence. A stochastic variant of the CVFR model is also studied and found to be robust to aversarial random attacks, which corrupt the items to be classified. This remarkable finding is one of the very many surprising effects which arise when noise and dynamical attributes are made to mutually resonate.