FILTER MODE ACTIVE

#active learning

Records found: 5

#active learning03/11/2025

Train High-Performance Supervised Models with Active Learning and Just a Few Labels

Active learning lets models query the most informative unlabeled samples for labeling, enabling high accuracy with far fewer annotations; this tutorial simulates a 20-query scenario using sklearn.

READ →

#active learning23/10/2025

FLAME: One-Step Active Learning for Lightning-Fast Remote Sensing Specialization

'FLAME adds a tiny, trainable refiner on top of an open-vocabulary detector to select the most informative samples and reach 53.96% AP on DOTA and 53.21% AP on DIOR with 30 labels and minute-scale CPU adaptation.'

READ →

#active learning11/10/2025

Master Self-Supervised Learning with Lightly: SimCLR, Coresets and Active Learning in Colab

'Step-by-step Colab tutorial showing how to train a SimCLR model with Lightly, visualize embeddings, select informative coresets and evaluate transfer performance with a linear probe.'

READ →

#active learning10/08/2025

From 100K to Under 500: Google’s Active Learning Cuts LLM Training Data by Orders of Magnitude

'Google shows active learning can fine-tune LLMs with under 500 targeted labels, matching or improving model quality while reducing labeling needs by orders of magnitude.'

READ →

#active learning11/05/2025

Accelerating Medical Symptom Classification with Adala and Google Gemini: A Hands-On Guide

This tutorial demonstrates building an active learning pipeline for medical symptom classification by integrating Adala with Google Gemini for efficient annotation and confidence visualization.

READ →