Learning to Generate with Memory
This addresses the challenge of capturing fine-grained details in generative models for applications such as image synthesis and data imputation, representing an incremental improvement over existing methods.
The paper tackles the problem of deep generative models losing local detail information during representation learning by introducing a model with external memory and smooth attention, achieving state-of-the-art results on tasks like density estimation and image generation.
Memory units have been widely used to enrich the capabilities of deep networks on capturing long-term dependencies in reasoning and prediction tasks, but little investigation exists on deep generative models (DGMs) which are good at inferring high-level invariant representations from unlabeled data. This paper presents a deep generative model with a possibly large external memory and an attention mechanism to capture the local detail information that is often lost in the bottom-up abstraction process in representation learning. By adopting a smooth attention model, the whole network is trained end-to-end by optimizing a variational bound of data likelihood via auto-encoding variational Bayesian methods, where an asymmetric recognition network is learnt jointly to infer high-level invariant representations. The asymmetric architecture can reduce the competition between bottom-up invariant feature extraction and top-down generation of instance details. Our experiments on several datasets demonstrate that memory can significantly boost the performance of DGMs and even achieve state-of-the-art results on various tasks, including density estimation, image generation, and missing value imputation.