Local Distance Preserving Auto-encoders using Continuous k-Nearest Neighbours Graphs
This work addresses representation learning for machine learning practitioners by improving local distance preservation in auto-encoders, though it appears incremental as it builds on existing auto-encoder frameworks.
The authors tackled the problem of learning representations that preserve local distances in auto-encoders by introducing models with a loss based on continuous k-nearest neighbours graphs and formulating training as a constraint optimization problem. Their method achieved state-of-the-art performance on several standard datasets and evaluation metrics.
Auto-encoder models that preserve similarities in the data are a popular tool in representation learning. In this paper we introduce several auto-encoder models that preserve local distances when mapping from the data space to the latent space. We use a local distance preserving loss that is based on the continuous k-nearest neighbours graph which is known to capture topological features at all scales simultaneously. To improve training performance, we formulate learning as a constraint optimisation problem with local distance preservation as the main objective and reconstruction accuracy as a constraint. We generalise this approach to hierarchical variational auto-encoders thus learning generative models with geometrically consistent latent and data spaces. Our method provides state-of-the-art performance across several standard datasets and evaluation metrics.