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

Multi-task learning (MTL) with neural networks leverages commonalities intasks to improve performance, but often suffers from task interference whichreduces the benefits of transfer. To address this issue we introduce therouting network paradigm, a novel neural network and training algorithm. Arouting network is a kind of self-organizing neural network consisting of twocomponents: a router and a set of one or more function blocks. A function blockmay be any neural network - for example a fully-connected or a convolutionallayer. Given an input the router makes a routing decision, choosing a functionblock to apply and passing the output back to the router recursively,terminating when a fixed recursion depth is reached. In this way the routingnetwork dynamically composes different function blocks for each input. Weemploy a collaborative multi-agent reinforcement learning (MARL) approach tojointly train the router and function blocks. We evaluate our model againstcross-stitch networks and shared-layer baselines on multi-task settings of theMNIST, mini-imagenet, and CIFAR-100 datasets. Our experiments demonstrate asignificant improvement in accuracy, with sharper convergence. In addition,routing networks have nearly constant per-task training cost while cross-stitchnetworks scale linearly with the number of tasks. On CIFAR-100 (20 tasks) weobtain cross-stitch performance levels with an 85% reduction in training time.

Routing Networks: Adaptive Selection of Non-linear Functions for Multi-Task Learning