The Kanerva Machine: A Generative Distributed Memory
This addresses the need for more efficient and scalable memory mechanisms in generative AI, though it is incremental as it builds on existing sparse distributed memory concepts.
The paper tackles the problem of quickly adapting generative models to new data by introducing an end-to-end trained memory system, which improves performance on Omniglot and CIFAR datasets and offers greater capacity and easier training compared to DNC variants.
We present an end-to-end trained memory system that quickly adapts to new data and generates samples like them. Inspired by Kanerva's sparse distributed memory, it has a robust distributed reading and writing mechanism. The memory is analytically tractable, which enables optimal on-line compression via a Bayesian update-rule. We formulate it as a hierarchical conditional generative model, where memory provides a rich data-dependent prior distribution. Consequently, the top-down memory and bottom-up perception are combined to produce the code representing an observation. Empirically, we demonstrate that the adaptive memory significantly improves generative models trained on both the Omniglot and CIFAR datasets. Compared with the Differentiable Neural Computer (DNC) and its variants, our memory model has greater capacity and is significantly easier to train.