← Back to AI Factory Chat

AI Factory Glossary

265 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 4 of 6 (265 entries)

neuron coverage, interpretability

Neuron coverage measures percentage of neurons activated by test inputs for test adequacy.

neuron-level analysis, explainable ai

Study individual neurons.

nevae, graph neural networks

NeVAE generates molecules by combining VAE with neural networks for structure generation.

never give up, ngu, reinforcement learning

Advanced exploration method.

never-ending learning,continual learning

System that learns indefinitely.

new product introduction, npi, business

Launch new product.

newsletter generation,content creation

Create email newsletters.

newsletter,blog,research,updates

Follow AI newsletters and research blogs to stay current. Field moves fast; continuous learning essential.

newsqa, evaluation

QA from news articles.

next sentence prediction, nsp, nlp

Predict if sentences are consecutive.

next token prediction,causal lm

Next token prediction is GPT-style training. Predict next token given previous. Autoregressive generation.

nextitnet, recommendation systems

NextItNet uses dilated convolutional neural networks for session-based recommendation capturing patterns at multiple temporal scales.

nextjs,react,fullstack

Next.js is React fullstack framework. SSR, API routes. Modern web apps.

nfnet, computer vision

High-performance nets without normalization.

ngu, ngu, reinforcement learning advanced

Never Give Up combines episodic and life-long curiosity to maintain exploration throughout training by tracking state visitation in embedding space.

nhwc layout, nhwc, model optimization

NHWC layout stores tensors in batch-height-width-channel order optimizing certain operations.

nickel contamination, contamination

Ni impurity effects.

nickel silicide (nisi),nickel silicide,nisi,feol

Alternative silicide material.

nisq (noisy intermediate-scale quantum),nisq,noisy intermediate-scale quantum,quantum ai

Current generation of quantum computers.

nisq era algorithms, nisq, quantum ai

Algorithms for noisy intermediate-scale quantum.

nist traceable,quality

Calibrated against US national standards.

nitridation,diffusion

Incorporate nitrogen into oxide to improve reliability.

nitride deposition,cvd

Grow Si3N4 using SiH4 + NH3 or similar.

nitrogen in silicon, material science

Nitrogen dopant or impurity.

nitrogen purge, packaging

Replace air with nitrogen.

nldm (non-linear delay model),nldm,non-linear delay model,design

Table-based timing model.

nlpaug,text,augmentation

nlpaug augments text data. Synonym, back-translation, contextual.

nlvr (natural language for visual reasoning),nlvr,natural language for visual reasoning,evaluation

Reason about image pairs.

nmf, nmf, recommendation systems

Non-negative Matrix Factorization constrains factors to be non-negative enabling interpretable part-based representations for recommendation.

no code,low code,build

No-code AI tools. Build without programming.

no-clean flux, packaging

Flux not requiring removal.

no-flow underfill, packaging

Apply before reflow.

no-repeat n-gram, llm optimization

No-repeat n-gram blocking prevents exact phrase repetition.

no-repeat n-gram, text generation

Prevent repeating n-grams.

no-u-turn sampler (nuts),no-u-turn sampler,nuts,statistics

Adaptive HMC algorithm.

node migration, business & strategy

Node migration transitions designs to newer process technologies.

node2vec, graph neural networks

Learn node embeddings via random walks.

noise augmentation, audio & speech

Noise augmentation adds background sounds to clean speech improving recognition in noisy conditions.

noise contrastive estimation for ebms, generative models

Train by contrasting data and noise distributions.

noise contrastive estimation, nce, machine learning

Learn by contrasting data with noise.

noise contrastive, structured prediction

Noise contrastive estimation trains unnormalized models by discriminating between data samples and artificially generated noise samples.

noise factors, doe

Uncontrollable variables causing variation.

noise floor, metrology

Minimum detectable signal.

noise multiplier, training techniques

Noise multiplier scales added noise controlling privacy-utility tradeoff.

noise schedule, generative models

Variance schedule for diffusion.

noisy labels learning,model training

Train robust models despite label errors.

noisy student, advanced training

Noisy student training iteratively generates pseudo-labels with a teacher model and trains larger student models with added noise.

nominal-the-best, quality & reliability

Nominal-the-best characteristics should achieve specific target values.

non-autoregressive generation, text generation

Generate all tokens at once.

non-autoregressive translation, nlp

Generate entire sequence in parallel.