AI Paper Atlas

Research map / Continual learning / Online and replay-based

Reparameterizing Convolutions for Incremental Multi-Task Learning without Task Interference

Menelaos Kanakis, David Bruggemann, Suman Saha, Stamatios Georgoulis, Anton Obukhov, Luc Van Gool

ECCV 2020 · Continual learning › Online and replay-based · approach: Dynamic expansion and isolation · 79 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
add response initialization (RI) (ablation)RC · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 2.13 -> 1.12Improved
add response initialization (RI) (ablation)RC+NFF · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 2.07 -> 0.99Improved
add normalized feature fusion (NFF) (ablation)RC · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 2.13 -> 2.07Improved
add normalized feature fusion (NFF) (ablation)RC+RI · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 1.12 -> 0.99Improved
replace shared convolutions with reparameterized convolutions (RC) (ablation)fully shared multi-task model · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 3.32 -> 2.13Improved
make BNs task-specific (ablation)frozen shared encoder · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 14.98 -> 5.76Improved
make Convs task-specific (ablation)frozen shared encoder · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 14.98 -> 0.62Improved
use RCM instead of Parallel RAtask-conditional MTL · PASCAL-Context (5 tasks)avg per-task performance drop vs single-task (Δm %): 2.09 -> 0.99Improved
use RCM instead of Parallel RAtask-conditional MTL · NYUD (4 tasks)avg per-task performance drop vs single-task (Δm %): 5.02 -> 1.48Improved
use RCM instead of ASTMT in incremental setupincremental learning · PASCAL-Context incremental (low-level tasks first)avg per-task performance drop vs single-task (Δm %): 11.77 -> 1.26Improved
use RCM instead of ASTMT in incremental setup (high-level tasks first)incremental learning · PASCAL-Context incremental (high-level tasks first)avg per-task performance drop vs single-task (Δm %): 3.76 -> 0.52Improved
use deeper ResNet-34 backbone instead of ResNet-18RCM · PASCAL-Context / NYUDavg per-task performance drop vs single-task (Δm %): 1.18 and 1.77No clear effect

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