AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Semantic Curiosity for Active Visual Learning

Devendra Singh Chaplot, Helen Jiang, Saurabh Gupta, Abhinav Gupta

ECCV 2020 · Reinforcement learning › Visual and representation RL · approach: Vision-based robot RL · 71 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 semantic curiosity exploration instead of random exploration to collect training datadetector fine-tuning · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 37.4 -> 39.96Improved
use semantic curiosity instead of Active Neural SLAM explorationdetector fine-tuning · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 38.5 -> 39.96Improved
use semantic curiosity instead of coverage explorationdetector fine-tuning · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 36.56 -> 39.96Improved
use semantic curiosity instead of object exploration rewarddetector fine-tuning · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 35.26 -> 39.96Improved
use prediction-error curiosity instead of random explorationdetector fine-tuning · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 37.4 -> 37.26No clear effect
fine-tune detector on random-policy trajectories vs not fine-tuningCOCO-pretrained detector · Matterport test (11 scenes), detector fine-tuned on sampled trajectoriesAP50 average: 31.72 -> 37.4Improved
use semantic curiosity instead of Active Neural SLAM for covering areaexploration policy · Replica trajectories (exploration analysis)Explored area: 11.527 -> 9.726Hurt
use semantic curiosity instead of object exploration reward for detections countexploration policy · Replica trajectories (exploration analysis)Number of object detections: 382.27 -> 291.78Hurt
use semantic curiosity instead of Active Neural SLAMexploration policy · Replica trajectories (exploration analysis)Semantic curiosity reward: 3.589 -> 4.378Improved
use semantic curiosity instead of Coverage Exploration (pretrained detector mAP50 on sampled data)exploration policy · Replica trajectories, pretrained detector (5 classes)mAP50 of pretrained detector: 51.02 -> 39.2Mixed

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