← Back to AI Factory Chat

AI Factory Glossary

9,967 technical terms and definitions

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Showing page 171 of 200 (9,967 entries)

structured perceptron, structured prediction

Structured perceptron extends the perceptron algorithm to structured prediction by updating on structured outputs rather than class labels.

structured pruning, model optimization

Structured pruning removes entire channels layers or blocks enabling hardware-efficient acceleration.

structured pruning,channel,head

Structured pruning removes entire channels or heads. Hardware-friendly. Less flexible than unstructured.

structured pruning,model optimization

Remove entire channels layers or attention heads.

structured representations, representation learning

Represent data with explicit structure.

structured svm, structured prediction

Structured Support Vector Machines extend SVM framework to structured outputs using max-margin training with structured loss functions.

stuck-at fault, advanced test & probe

Stuck-at faults model defects where circuit nodes are permanently fixed at logic zero or one independent of circuit inputs.

stuck-at fault,testing

Signal permanently stuck at 0 or 1.

stuck-open fault, advanced test & probe

Stuck-open faults represent transistors failing to conduct detected through elevated IDDQ or two-pattern testing.

student teacher,smaller model,kd

Student is smaller model learning from larger teacher. Teacher provides richer signal than hard labels.

student-teacher framework for self-supervised, self-supervised learning

EMA teacher for student training.

study,learn,tutor

AI tutoring explains concepts. Personalized pace, examples.

stumps, stumps, design & verification

Self-Test Using MISR and Parallel SRSG provides BIST architecture.

style loss, generative models

Match style statistics.

style mixing, generative models

Combine styles from different images.

style mixing, multimodal ai

Style mixing combines latent codes at different scales creating hybrid generations.

style reference, generative models

Match style of reference image.

style transfer diffusion, multimodal ai

Style transfer in diffusion models adapts content to reference styles through conditioning.

style transfer,generative models

Apply artistic style from one image to content of another.

style-based generation,generative models

Generate content in specific artistic styles.

style,tone,format

Style instructions control output: formal/casual, brief/detailed, bullet/prose. Customize communication.

stylegan architecture, generative models

Style-based generator.

stylegan-xl,generative models

Large-scale StyleGAN.

stylegan,generative models

High-quality GAN with style control.

stylegan3, multimodal ai

StyleGAN3 improves alias-free generation through rotation and translation equivariance.

stylus profilometer,metrology

Mechanical surface roughness measurement.

sub-question decomposition, rag

Break into multiple queries.

subgoal, ai agents

Subgoals are intermediate objectives required to achieve overall goals.

subgradient method, structured prediction

Subgradient methods optimize structured prediction objectives by computing subgradients of non-smooth loss functions.

subgraph isomorphism, graph algorithms

Determine if one graph contains another.

subgroup frequency, quality & reliability

Subgroup frequency determines sampling rate for detecting process shifts.

subgroup size, quality & reliability

Subgroup size affects control chart sensitivity and capability estimation accuracy.

subject-driven generation, multimodal ai

Subject-driven generation creates images featuring specific subjects from reference images.

subsampling, training techniques

Subsampling randomly selects training examples reducing privacy cost.

subspace alignment, domain adaptation

Align domain subspaces.

substitutional impurity, defects

Atom replacing lattice atom.

substrate noise coupling,design

Unwanted signals through substrate.

subthreshold computing,design

Operate below threshold voltage.

subtractive etch beol, process integration

Subtractive etch patterns metal by depositing then etching rather than damascene filling.

success run theorem, reliability

Statistical inference from zero failures.

success testing, reliability

Demonstrate reliability goal.

successive inspection, quality & reliability

Successive inspection verifies output before next operation begins.

successor features, reinforcement learning advanced

Successor features decouple environment dynamics from rewards enabling transfer across tasks with shared state space.

successor representation, reinforcement learning advanced

Successor representation separates environment dynamics from rewards enabling fast policy evaluation under new rewards.

suggestion system, quality & reliability

Suggestion systems encourage employee ideas for improvement.

summarization for context, prompting

Summarize old messages to save space.

summarization,compress,distill

Summarization compresses long text to key points. Use for fitting long docs into context or creating memory.

summarize,condense,tldr

Summarize long documents. Key points extraction.

summary generation as pre-training, nlp

Generate summaries during pre-training.

sundae, sundae, text generation

Stepwise refinement for text generation.