TACOS: Task Agnostic Continual Learning in Spiking Neural Networks
This addresses the problem of catastrophic interference for AI systems in real-world scenarios where task boundaries are unknown, though it is incremental as it builds on bio-inspired approaches.
The paper tackles catastrophic interference in machine learning by introducing TACOS, a spiking neural network model that uses neuro-inspired mechanisms like synaptic consolidation and metaplasticity to mitigate memory loss without task awareness, outperforming existing regularization techniques in domain-incremental learning scenarios.
Catastrophic interference, the loss of previously learned information when learning new information, remains a major challenge in machine learning. Since living organisms do not seem to suffer from this problem, researchers have taken inspiration from biology to improve memory retention in artificial intelligence systems. However, previous attempts to use bio-inspired mechanisms have typically resulted in systems that rely on task boundary information during training and/or explicit task identification during inference, information that is not available in real-world scenarios. Here, we show that neuro-inspired mechanisms such as synaptic consolidation and metaplasticity can mitigate catastrophic interference in a spiking neural network, using only synapse-local information, with no need for task awareness, and with a fixed memory size that does not need to be increased when training on new tasks. Our model, TACOS, combines neuromodulation with complex synaptic dynamics to enable new learning while protecting previous information. We evaluate TACOS on sequential image recognition tasks and demonstrate its effectiveness in reducing catastrophic interference. Our results show that TACOS outperforms existing regularization techniques in domain-incremental learning scenarios. We also report the results of an ablation study to elucidate the contribution of each neuro-inspired mechanism separately.