A-Z of Machine Learning and Computer Vision Terms
A
Activation Function
Active Learning
AI Agent
AI Assistants
AI Assisted Labeling
Algorithm
Anchor Boxes
Anomaly Detection
Artificial Intelligence (AI)
Attribute
B
Backpropagation
Backpropagation
Bagging
Batch
Batch Normalization
Bayesian Network
Bias
Big Data
Binary Classification
Blur
Boosting
Bounding Box
C
Calibration
Calibration Curve
Canonical Correlation Analysis (CCA)
Case-Based Reasoning
Chain of Thought (CoT)
ChatGPT
Chi-Squared Automatic Interaction Detection (CHAID)
Class Boundary (Statistics & Machine Learning)
Class Imbalance
Clustering
COCO
Collaborative Filtering
Computer Vision
Computer Vision Model
Concept Drift
Conditional Random Field (CRF)
Confusion Matrix
Constrained Clustering
Contrastive Learning
Convolutional Neural Network (CNN)
Cross-entropy Loss
Cross-Validation
D
Data Approximation
Data Augmentation
Data Drift
Data Error
Data Mining
Data Operations
Data Pre-processing
Data Quality
Dataset
Decision Boundary
Decision List
Decision Stump
Decision Tree
Deep Learning
Deep Neural Networks
Deep Reinforcement Learning (DRL)
DICOM
Dimensionality Reduction
Dimensionality Reduction
Dropout
Dynamic and Event-Based Classifications
E
Edge Cases
Edge Computing
Edge Detection
Elastic Net
Embedding Spaces
Ensemble Learning
Epoch
Expectation-Maximization Algorithm (EM)
Extreme Learning Machine
F
F1 Score
Factor Analysis
False Positive Rate
Feature
Feature Extraction
Feature Hashing
Feature Learning
Feature Scaling
Feature Selection
Feature Vector
Fisher’s Linear Discriminant
Foundation Models
FP-Growth Algorithm
Frame Rate
Frames Per Second (FPS)
Fully Connected Layer
Fuzzy Logic
G
Generative Adversarial Network (GAN)
Generative Adversarial Networks
Pool-Based Sampling
Pool-based sampling is an active learning strategy where a model is given access to a large pool of unlabeled data and must choose the most informative samples from that pool to be labeled next​. The model (or an acquisition function) typically selects examples about which it is uncertain (or which maximize some information criteria) from the pool. These selected samples are then annotated by an oracle (e.g., a human), added to the training set, and the model is retrained​. This iterative process focuses labeling efforts on the data that will most improve the model, reducing the total number of annotations needed. Common pool-based methods include uncertainty sampling, query-by-committee, and density-weighted sampling​
Explore Our Products
.svg)
LightlyStudio
Data Curation, Labeling & QA
Get data insights and find the perfect
selection strategy, label &Â QA data.
LightlyTrain
Self-Supervised Pretraining &Â Fine-Tuning
Leverage self-supervised learning to pretrain
and fine-tune models
