English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

While normalizing flows for continuous data have been extensively researched,flows for discrete data have only recently been explored. These prior models,however, suffer from limitations that are distinct from those of continuousflows. Most notably, discrete flow-based models cannot be straightforwardlyoptimized with conventional deep learning methods because gradients of discretefunctions are undefined or zero. Previous works approximate pseudo-gradients ofthe discrete functions but do not solve the problem on a fundamental level. Inaddition to that, backpropagation can be computationally burdensome compared toalternative discrete algorithms such as decision tree algorithms. Our approachseeks to reduce computational burden and remove the need for pseudo-gradientsby developing a discrete flow based on decision trees -- building upon thesuccess of efficient tree-based methods for classification and regression fordiscrete data. We first define a tree-structured permutation (TSP) thatcompactly encodes a permutation of discrete data where the inverse is easy tocompute; thus, we can efficiently compute the density value and sample newdata. We then propose a decision tree algorithm to build TSPs that learns thetree structure and permutations at each node via novel criteria. We empiricallydemonstrate the feasibility of our method on multiple datasets.

Discrete Tree Flows via Tree-Structured Permutations