Bayesian Continual Learning via Spiking Neural Networks
This work addresses the problem of enabling continual learning and risk management in neuromorphic computing, which is incremental by building on existing Bayesian and spiking neural network frameworks.
The paper tackled the challenge of designing neuromorphic systems that adapt to changing tasks while providing uncertainty estimates, by deriving online Bayesian learning rules for spiking neural networks. Experimental results on Intel's Lava platform demonstrated improved adaptation and uncertainty quantification compared to frequentist methods.
Among the main features of biological intelligence are energy efficiency, capacity for continual adaptation, and risk management via uncertainty quantification. Neuromorphic engineering has been thus far mostly driven by the goal of implementing energy-efficient machines that take inspiration from the time-based computing paradigm of biological brains. In this paper, we take steps towards the design of neuromorphic systems that are capable of adaptation to changing learning tasks, while producing well-calibrated uncertainty quantification estimates. To this end, we derive online learning rules for spiking neural networks (SNNs) within a Bayesian continual learning framework. In it, each synaptic weight is represented by parameters that quantify the current epistemic uncertainty resulting from prior knowledge and observed data. The proposed online rules update the distribution parameters in a streaming fashion as data are observed. We instantiate the proposed approach for both real-valued and binary synaptic weights. Experimental results using Intel's Lava platform show the merits of Bayesian over frequentist learning in terms of capacity for adaptation and uncertainty quantification.