Park, M;
Jitkrittum, W;
Qamar, A;
Szabo, Z;
Buesing, L;
Sahani, M;
(2014)
Bayesian Manifold Learning: The Locally Linear Latent Variable Model (LL-LVM).
In:
Advances in Neural Information Processing Systems 28 (NIPS 2015).
Neural Information Processing Systems Foundation: Montreal, Canada.
Preview |
Text
Szabo_5973-bayesian-manifold-learning-the-locally-linear-latent-variable-model-ll-lvm.pdf Download (2MB) | Preview |
Abstract
We introduce the Locally Linear Latent Variable Model (LL-LVM), a probabilistic model for non-linear manifold discovery that describes a joint distribution over observations, their manifold coordinates and locally linear maps conditioned on a set of neighbourhood relationships. The model allows straightforward variational optimisation of the posterior distribution on coordinates and locally linear maps from the latent space to the observation space given the data. Thus, the LL-LVM encapsulates the local-geometry preserving intuitions that underlie non-probabilistic methods such as locally linear embedding (LLE). Its probabilistic semantics make it easy to evaluate the quality of hypothesised neighbourhood relationships, select the intrinsic dimensionality of the manifold, construct out-of-sample extensions and to combine the manifold model with additional probabilistic models that capture the structure of coordinates within the manifold.
Archive Staff Only
 |
View Item |
