Lifelong Neural Predictive Coding: Learning Cumulatively Online without Forgetting

In lifelong learning systems, especially those based on artificial neural networks, one of the biggest obstacles is the severe inability to retain old knowledge as new information is encountered. This phenomenon is known as catastrophic forgetting. In this article, we propose a new kind of connectionist architecture, the Sequential Neural Coding Network, that is robust to forgetting when learning from streams of data points and, unlike networks of today, does not learn via the immensely popular back-propagation of errors. Grounded in the neurocognitive theory of predictive processing, our model adapts its synapses in a biologically-plausible fashion, while another, complementary neural system rapidly learns to direct and control this cortex-like structure by mimicking the task-executive control functionality of the basal ganglia. In our experiments, we demonstrate that our self-organizing system experiences significantly less forgetting as compared to standard neural models and outperforms a wide swath of previously proposed methods even though it is trained across task datasets in a stream-like fashion. The promising performance of our complementary system on benchmarks, e.g., SplitMNIST, Split Fashion MNIST, and Split NotMNIST, offers evidence that by incorporating mechanisms prominent in real neuronal systems, such as competition, sparse activation patterns, and iterative input processing, a new possibility for tackling the grand challenge of lifelong machine learning opens up.