AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Extensions of Karger's Algorithm: Why They Fail in Theory and How They Are Useful in Practice

Erik Jenner, Enrique Fita Sanmartín, Fred A. Hamprecht

ICCV 2021 · Label-efficient and robust learning › Semi-supervised learning · approach: Holistic semi-supervised pipelines · 2 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 Karger-contraction potential instead of random walker potentialseeded label propagation · USPS semi-supervised (20 labeled seeds)accuracy (%): 53.7 -> 62.2Improved
use Karger-contraction potential instead of random walker potentialseeded label propagation · USPS semi-supervised (40 labeled seeds)accuracy (%): 68.0 -> 73.1Improved
use Karger-contraction potential instead of random walker potentialseeded label propagation · USPS semi-supervised (100 labeled seeds)accuracy (%): 87.7 -> 89.0Improved
use Karger-contraction potential instead of random walker potentialseeded label propagation · USPS semi-supervised (200 labeled seeds)accuracy (%): 92.4 -> 92.6No clear effect
use Karger-contraction potential instead of random walker potentialseeded segmentation · Grabcut (seeded segmentation)accuracy (%): 96.0 -> 96.3No clear effect
use Karger-contraction potential instead of random walker potentialseeded segmentation · Grabcut (seeded segmentation)ARI: 0.82 -> 0.82No clear effect
use Karger-contraction potential instead of random walker potentialseeded segmentation · Grabcut (seeded segmentation)VoI: 0.25 -> 0.24No clear effect
use random walker instead of watershedseeded label propagation · USPS semi-supervised (100 labeled seeds)accuracy (%): 74.3 -> 87.7Improved
use Power Watershed instead of watershedseeded label propagation · USPS semi-supervised (200 labeled seeds)accuracy (%): 80.0 -> 80.8No clear effect
use Power Watershed instead of watershedseeded segmentation · Grabcut (seeded segmentation)ARI: 0.83 -> 0.83No 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