AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

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

Clemens Rosenbaum, Tim Klinger, Matthew Riemer

ICLR 2018 · Architectures and efficiency › RNNs and normalization · approach: Mechanistic circuit interpretability · 35 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use per-task hard routing trained with WPL instead of cross-stitch sharingSimpleConvNet · CIFAR-MTLaccuracy: 60 -> 53Improved
use per-task hard routing trained with WPL instead of cross-stitch sharingSimpleConvNet · MIN-MTLaccuracy: 57 -> 55Improved
use per-task hard routing trained with WPL instead of task-specific fully connected layersSimpleConvNet · MNIST-MTLaccuracy: 99 -> 96Improved
use WPL rather than tabular Q-learningCIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 55Improved
use multiple agents, one per task, rather than a single agent (ablation)CIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 44Improved
use a tabular policy instead of an approximate policy (ablation)CIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 25Improved
add a learned dispatcher to assign instances to agents (ablation)CIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 41Hurt
replace hard routing with soft mixture routing (ablation)CIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 35Hurt
allow non-layered routing with PASS (ablation)CIFAR-MTL routing network · CIFAR-MTLaccuracy: 60 -> 46Hurt
add router-choice gradients to the function-block gradients (ablation)routing network training · classification experimentsperformanceNo clear effect
use average historical block-choice probability rather than average historical choice count for the collaboration reward (ablation)routing network router · routing experimentsfinal performanceNo clear effect
set collaboration reward to zero or a small value (ablation)routing network router · CIFAR-MTL and MIN-MTLaccuracyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy