active-outlier-detection machine learning
Active Learning for Outlier Detection Introduction & Motivation Active Learning for Outlier Detection addresses a central problem in {subject} learning and efficient {target}…
28,973 technical terms and definitions
Active Learning for Outlier Detection Introduction & Motivation Active Learning for Outlier Detection addresses a central problem in {subject} learning and efficient {target}…
Active Preference Elicitation via Pairwise Comparisons Introduction & Motivation Active Preference Elicitation via Pairwise Comparisons addresses a central problem in {subject}…
prompting techniques
Active Prompting is an adaptive prompting approach that focuses additional effort on uncertain or difficult queries - It is a core method in modern LLM execution workflows. What…
Active Prototype and Example Selection Introduction & Motivation Active Prototype and Example Selection addresses a central problem in {subject} learning and efficient {target}…
forward active mode, BJT active region, linear amplification, Early effect, collector current, beta, MOSFET saturation caveat
The active region, also called the forward-active mode, is the operating condition of a bipolar junction transistor in which the base-emitter junction is forward biased and…
rag
Active retrieval is the adaptive retrieval policy where the model decides when and what to retrieve during reasoning rather than using a fixed one-shot fetch - it aligns retrieval…
Active Search and Rare Category Discovery Introduction & Motivation Active Search and Rare Category Discovery addresses a central problem in {subject} learning and efficient…
Constrained Seeded K-Means with Must-Link Cannot-Link Introduction & Motivation Constrained Seeded K-Means with Must-Link Cannot-Link addresses a central problem in semisupervised…
Graph-Based Label Propagation Active Learning Introduction & Motivation Graph-Based Label Propagation Active Learning addresses a central problem in {subject} learning…
Active Learning in Semi-Supervised Setting Introduction & Motivation Active Learning in Semi-Supervised Setting addresses a central problem in semisupervised…
model optimization
Active Shift is a learnable shift mechanism where displacement parameters are optimized during training - It extends fixed shift operations with adaptive spatial routing. What Is…
Active Testing for Label-Efficient Model Evaluation Introduction & Motivation Active Testing for Label-Efficient Model Evaluation addresses a central problem in {subject} learning…
Active Domain Adaptation via Target Margin Sampling Introduction & Motivation Active Domain Adaptation via Target Margin Sampling addresses a central problem in {subject} learning…
Active Learning for Transfer Tasks Introduction & Motivation Active Learning for Transfer Tasks addresses a central problem in {subject} learning and efficient {target}…
Activity Localization in Video Introduction & Motivation Activity Localization in Video addresses a central problem in multimodal learning and temporal-localization-accuracy…
quality & reliability
Activity Network is a dependency map that sequences project activities and reveals logical execution flow - It is a core method in modern semiconductor quality governance…
activity recognition, sensor data analysis, cnn har lstm ntu passed rgb, activity recognition & sensor data analysis, activity recognition sensor, recognition sensor data, recognition sensor
Activity Recognition & Sensor Data Analysis Introduction & Motivation Activity Recognition: classify human actions from sensor data. Accelerometer, gyroscope, inertial sensors.…
activity selection greedy, activity selection, selection greedy
Activity Selection Greedy Introduction & Motivation Activity Selection Greedy addresses a central problem in algorithms process control and activity-count-maximum prediction: how…
101240 actor critic methods, 101240 actor, actor critic, a3c ai ddpg gae td, batch normalization & layer normalization, batch normalization layer, normalization layer normalization, batch normalization
Batch Normalization & Layer Normalization Introduction & Motivation Normalization: stabilize training. Batch norm, layer norm, instance norm. Applications: deeper networks, faster…
actor critic methods a3c sac hybrid algorithms, actor critic, methods a3c sac, a2c a3c ai dqn lr passed, actor-critic methods a3c sac & hybrid algorithms, actor-critic methods a3c, methods a3c
Actor-Critic Methods: A3C, SAC & Hybrid Algorithms Introduction & Motivation Actor-critic: combine policy (actor) and value (critic) functions. A3C: asynchronous advantage…
Actor-Critic Learning Algorithms Introduction & Motivation Actor-Critic Learning Algorithms addresses a central problem in reinforcement policy learning…
erlang actor, akka actor, message passing actor, actor framework
The Actor Model is the concurrent programming paradigm where the fundamental unit of computation is the actor — an isolated entity that communicates exclusively through…
actuator control motor, actuator control, control motor
Actuator Control Motor Introduction & Motivation Actuator Control Motor addresses a central problem in robotics process control and motor-control-speed-torque prediction: how…
marketing, persuade
AI advertising copy generation creates persuasive marketing content at scale. Copywriting frameworks: AIDA (Attention, Interest, Desire, Action), PAS (Problem, Agitation…
content creation
Ad creative generation is the use of AI to automatically produce advertisement visuals, copy, and multimedia content — creating complete ad assets including images, videos, text…
ada lovelace 1843, first computer program, first computer algorithm, note g bernoulli numbers, analytical engine program, lovelace babbage, luigi menabrea, history of programming
Augusta Ada King, Countess of Lovelace, born in 1815 as the only legitimate daughter of the poet Lord Byron, spent 1842 and 1843 translating a French memoir on Babbage's unbuilt…
optimization
AdaBelief is an adaptive optimizer that adapts the learning rate based on the "belief" in the current gradient direction — using the deviation of the gradient from the expected…
adaptive, weight
AdaBoost (Adaptive Boosting) is the original boosting algorithm that combines many "weak learners" (typically decision stumps — single-split trees) into a powerful ensemble…
adaboost adaptive boosting prediction, adaboost adaptive boosting, adaptive boosting prediction, adaboost adaptive, adaptive boosting, boosting prediction
AdaBoost Adaptive Boosting Prediction with Machine Learning Introduction & Motivation AdaBoost Adaptive Boosting Prediction with Machine Learning addresses a central problem…
AdaDelta Dimensionless Adaptive Learning Rate Introduction & Motivation AdaDelta Dimensionless Adaptive Learning Rate addresses a central problem in optimization algorithm tuning…
optimization
Adafactor is a memory-efficient adaptive optimizer designed for training large models — replacing Adam's per-parameter second moment buffer with a factored approximation, reducing…
adagrad accumulated gradient scaling and adaptive features, adagrad accumulated gradient, accumulated gradient scaling, gradient scaling adaptive, scaling adaptive features
AdaGrad Accumulated Gradient Scaling and Adaptive Features Introduction & Motivation AdaGrad Accumulated Gradient Scaling and Adaptive Features addresses a central problem…
AdaGrad Adaptive Gradient Subgradient Scaling Introduction & Motivation AdaGrad Adaptive Gradient Subgradient Scaling addresses a central problem in optimization algorithm tuning…
adamw, optimizer, weight decay, training, lr, momentum
An optimizer is the rule that turns gradients into weight updates. Backpropagation tells you the direction of steepest descent for every parameter; the optimizer decides how far…
model training
Adam combines momentum and per-parameter adaptive scaling into a single update rule, tracking two running averages for every weight: a first moment (the mean of recent gradients…
adamw, rmsprop, optimizer comparison
An optimizer is the rule that turns gradients into weight updates. Backpropagation tells you the direction of steepest descent for every parameter; the optimizer decides how far…
adam optimizer adaptive learning rate and moment estimation, adam optimizer adaptive, optimizer adaptive learning, adaptive learning rate, learning rate moment, rate moment
Adam Optimizer Adaptive Learning Rate and Moment Estimation Introduction & Motivation Adam Optimizer Adaptive Learning Rate and Moment Estimation addresses a central problem…
Adam Optimizer Parameter Tuning Introduction & Motivation Adam Optimizer Parameter Tuning addresses a central problem in optimization algorithm tuning and improvement: how…
model training
An optimizer is the rule that turns gradients into weight updates. Backpropagation tells you the direction of steepest descent for every parameter; the optimizer decides how far…
AdamW Optimizer with Decoupled Weight Decay Introduction & Motivation AdamW Optimizer with Decoupled Weight Decay addresses a central problem in optimization algorithm tuning…
computer vision
AdamW is the universally adopted, mathematically corrected optimizer for training Vision Transformers and all modern Transformer-based architectures — critically fixing…
adapter modules, adapter layers, Houlsby 2019, bottleneck adapter, AdapterFusion, AdapterHub, parameter-efficient transfer learning
An adapter is a small trainable module inserted into a pretrained neural network so that the network can learn a new task while its original weights stay frozen. The idea was…
continual learning
Adapter-based continual learning is continual learning that adds lightweight adapter modules for each new task instead of retraining full models - Adapters isolate task updates…
AdapterFusion for Dynamic Multi-Task Knowledge Transfer Introduction & Motivation AdapterFusion for Dynamic Multi-Task Knowledge Transfer addresses a central problem in transfer…
fine-tuning
Adapter layers are small trainable modules inserted into frozen pretrained models, enabling parameter-efficient fine-tuning by learning task-specific transformations without…
adapter fine tuning, lora, houlsby adapter, parallel adapter, parameter efficient fine tuning, peft
Adapter tuning customizes a frozen or mostly frozen pretrained model by training small task-specific parameter modules. It lowers optimizer memory, communication, storage…
learnable activation, prelu
Adaptive activation refers to activation functions whose shape or parameters are learned during training rather than fixed in advance. This makes the network more flexible because…
design
Adaptive aging compensation is the runtime control strategy that counteracts performance loss from device aging using monitored silicon condition - it adjusts voltage, frequency…
ai safety
Adaptive Attacks are adversarial attacks specifically designed to overcome a particular defense mechanism — tailoring the attack strategy to exploit the defense's specific…
Adaptive Dynamic Batch Sizing in Active Learning Introduction & Motivation Adaptive Dynamic Batch Sizing in Active Learning addresses a central problem in {subject} learning…