Search Results for author:
Found 10 papers, 1 papers with code
Finito: A Faster, Permutable Incremental Gradient Method for Big Data Problems
Recent advances in optimization theory have shown that smooth strongly convex finite sums can be minimized faster than by treating them as a black box "batch" problem.
Learning as MAP Inference in Discrete Graphical Models
Instead of relying on convex losses and regularisers such as in SVMs, logistic regression and boosting, or instead non-convex but continuous formulations such as those encountered in neural networks and deep belief networks, our framework entails a non-convex but \emph{discrete} formulation, where estimation amounts to finding a MAP configuration in a graphical model whose potential functions are low-dimensional discrete surrogates for the misclassification loss.
A Convex Formulation for Learning Scale-Free Networks via Submodular Relaxation
We consider the case where the structure of the graph to be reconstructed is known to be scale-free.
Submodular Multi-Label Learning
The key novelty of our formulation is that we explicitly allow for assortative (submodular) pairwise label interactions, i. e., we can leverage the co-ocurrence of pairs of labels in order to improve the quality of prediction.
Word Features for Latent Dirichlet Allocation
We extend Latent Dirichlet Allocation (LDA) by explicitly allowing for the encoding of side information in the distribution over words.
Reverse Multi-Label Learning
Multi-label classification is the task of predicting potentially multiple labels for a given instance.
Multitask Learning without Label Correspondences
We propose an algorithm to perform multitask learning where each task has potentially distinct label sets and label correspondences are not readily available.
Exponential Family Graph Matching and Ranking
We present a method for learning max-weight matching predictors in bipartite graphs.
Convex Relaxation of Mixture Regression with Efficient Algorithms
The second is a min-min reformulation consisting of fast alternating steps of closed-form updates.
Robust Near-Isometric Matching via Structured Learning of Graphical Models
Models for near-rigid shape matching are typically based on distance-related features, in order to infer matches that are consistent with the isometric assumption.
