Alignment (Alignment)
Alignment
Sɛnea wɔyɛ no sɛ AI nhyehyɛe bi atirimpɔw, nea ɛma ɛba, ne ne suban ne nnipa atirimpɔw ne gyinapɛn hyia. Eyi ho hia titiriw wɔ nhyehyɛe a ɛkɔ anim a ebetumi anya suban a wɔammɔ ho kyerɛwtohɔ pɔtee.
Nhwɛso: Sɛnea wɔyɛ no sɛ chatbot a wɔde boa adwenemyarefo nkyerɛkyerɛ nneɛma a ɛyɛ hu da biara da, sɛnea ɛbɛyɛ a wɔbɛtumi ayɛ adwuma.
Application Programming Interface (API) (Application Programming Interface (API))
Application Programming Interface (API)
Mmara ne nhyehyɛe a wɔakyerɛkyerɛ mu a ɛma software nhyehyɛe ahorow tumi ne wɔn ho wɔn ho kasa na wɔsesa data.
Nhwɛso: Sɛ wode OpenAI API di dwuma de soma nsɛm na wunya kasa nhwɛso a wɔayɛ no wɔ wo wɛbsaet so.
Artificial General Intelligence (AGI) (Artificial General Intelligence (AGI))
Artificial General Intelligence (AGI)
AI kwan bi a É›wÉ” adwene mu a ebetumi ayÉ› adwuma biara a É›hwehwÉ› nnipa adwene. Æma adesua kÉ” anim wÉ” nsÉ›m ahorow mu.
Nhwɛso: AGI nhyehyɛe bi betumi asua nnwom a wɔyɛ, ayɛ oprehyɛn, na watumi apasa nyansapɛ sɔhwɛ a wonni adwuma pɔtee biara.
Artificial Intelligence (AI) (Artificial Intelligence (AI))
Artificial Intelligence (AI)
Nnipa adwene a wɔde yɛ mfidie a wɔayɛ no sɛnea ɛbɛyɛ a wɔbɛtumi adwen, abɔ nsusuwii, na wɔayɛ adwuma wɔ wɔn ankasa.
Nhwɛso: AI na ɛma nnipa boafo te sɛ Siri ne kar a ɛkɔ ne ho te sɛ Tesla Autopilot yɛ adwuma.
AI Ethics (AI Ethics)
AI Ethics
Nkyerɛkyerɛmu a ɛfa AI nkɔso ne ne dwumadi ho, a ɛfa trenee, ankorankoro nsɛm, akontabuo, ne nyiyim a enni ho.
Nhwɛso: Sɛnea wɔyɛ nhyehyɛe a ɛbɛsiw adwumayɛfo a wɔde wɔn di dwuma no kwan sɛ wɔnyɛ nyiyim wɔ ɔbarima ne ɔbea nna anaa mmusuakuw ho.
Augmented Intelligence (Augmented Intelligence)
Augmented Intelligence
Nkitahodi nhwɛso bi a AI boa na ɛma nnipa adwene kɔ anim sen sɛ ɛbɛsesa no.
Nhwɛso: AI-powered radiology nnwinnade a ɛkyerɛ nneɛma a ɛnteɛ ma adɔkotafo, a wɔyɛ gyinaesi a etwa to.
Autonomous Agent (Autonomous Agent)
Autonomous Agent
AI nhyehyɛe bi a ɛwɔ tumi sɛ ɛyɛ ne gyinaesi ankasa na ɛyɛ adwuma de du ne atirimpɔw ho a nnipa mfa wɔn ho nnye mu.
Nhwɛso: Kar a ɛkɔ ne ho a ɛde nneɛma kɔ a ɛfa kurow no mmɔnten so na ɛkwati nneɛma a ɛsiw kwan wɔ ne ho.
Backpropagation (Backpropagation)
Backpropagation
Akwan a wɔfa so kyerɛkyerɛ neural networks mu denam nneɛma a ɛyɛ den a wɔsesa no fi nea ɛba kosi nea ɛde ba, na ɛma nsɛm a wɔka no so tew.
Nhwɛso: Wɔde di dwuma wɔ mfonini a wɔde kyerɛkyerɛ mu de tew mfomso a ɛba wɔ nkyerɛwde a wɔde nsa kyerɛw no so.
Bias (Algorithmic Bias) (Bias (Algorithmic Bias))
Bias (Algorithmic Bias)
AI nea ɛba mu a ɛnyɛ nea wɔahyɛ da ayɛ na ɛyɛ nhyehyɛe a ɛma obiara nya mfaso esiane data a ɛnteɛ anaa nea ɛnyɛ nea ɛfata nti.
Nhwɛso: Aniwa a ɛhwehwɛ nnipa mu nhyehyɛe a ɛtaa hu nnipa a wɔyɛ kɔlafo mfomso esiane sɛ wɔanhyɛ wɔn ananmu wɔ nkyerɛkyerɛmu data mu nti.
Big Data (Big Data)
Big Data
Data kɛse a ɛhwehwɛ nnwinnade pɔtee de kora, hwehwɛ mu, na woyi nea ɛsom bo fi mu, a wɔtaa de kyerɛkyerɛ AI nhwɛso ahorow.
Nhwɛso: Sɛ wode nnipa ɔpepem pii nkitahodi di dwuma de kyerɛkyerɛ nhyehyɛe a ɛma wɔde nneɛma kyerɛ ma e-commerce platforms.
Black Box Model (Black Box Model)
Black Box Model
AI anaa machine learning nhwɛso bi a ne mu nsusuwii mu yɛ den ma nnipa sɛ wɔbɛkyerɛkyerɛ mu, na ɛma ɛyɛ den sɛ wɔbɛte sɛnea wɔyɛ gyinaesi ase.
Nhwɛso: Deep neural network a wɔde gye bosea nanso ɛmfa nkyerɛkyerɛmu a emu da hɔ mma nea enti a wɔagye ɔbɔfo bi na wɔapo ɔfoforo.
Cognitive Computing (Cognitive Computing)
Cognitive Computing
AI nhyehyɛe a wɔayɛ no sɛnea ɛbɛyɛ a ɛbɛyɛ nnipa adwene, te sɛ nsusuwii ne adesua, denam akwan te sɛ NLP ne nhwɛso a wɔde hu nneɛma so.
Nhwɛso: Cognitive computing nhyehyɛe bi a ɛboa mmara ho adwumayɛfo ma wɔhwehwɛ asɛm mu mmara mu na wɔka nea ebetumi aba ho asɛm.
Computer Vision (Computer Vision)
Computer Vision
Artificial intelligence fã bi a ɛma kɔmputa tumi kyerɛkyerɛ na ɛyɛ mfonini ne video data.
Nhwɛso: Aniwa a ɛhwehwɛ nnipa mu nhyehyɛe a ɛhu nnipa wɔ ahobammɔ mfonini mu denam kɔmputa aniwa so.
Corpus (Corpus)
Corpus
Nkyerɛwde anaa kasa kɛse a wɔde kyerɛkyerɛ kasa nhwɛso ahorow.
Nhwɛso: Common Crawl data set yɛ wɛb corpus a wɔde kyerɛkyerɛ kasa nhwɛso akɛse te sɛ GPT.
Data Drift (Data Drift)
Data Drift
Sɛnea input data sesa bere kɔ so, na ɛma nhwɛso no yɛ adwuma bɔne.
Nhwɛso: Nhwɛso a ɛka sɛ wɔbɛyɛ adwuma wɔ mfidie mu no yɛ adwuma bɔne bere a wɔde sensor foforo ba.
Data Labelling (Data Labelling)
Data Labelling
Sɛnea wɔde tags anaa labels kyerɛkyerɛ data mu de ma ɛfata sɛ wɔde di dwuma wɔ adesua a wɔhwɛ so mu.
Nhwɛso: Sɛnea wɔde labels kyerɛkyerɛ akisikuru mfonini mpempem pii mu sɛ ɛyɛ benign anaa malignant de kyerɛkyerɛ akisikuru a wɔde hu no mu.
Data Mining (Data Mining)
Data Mining
Sɛnea wɔde hu nhwɛso, nkitahodi, ne nneɛma a ɛnteɛ wɔ data kɛse mu.
Nhwɛso: Retailers a wɔde data mining di dwuma de hu sɛ nnipa a wɔtɔ nappies taa tɔ beer nso.
Deep Learning (Deep Learning)
Deep Learning
Machine learning fã bi a ɛde multi-layered neural networks di dwuma de yɛ nhwɛso a ɛyɛ den wɔ data mu.
Nhwɛso: Deep learning na wɔde di dwuma wɔ kasa nhwɛso ahorow te sɛ GPT-4 ne mfonini a wɔde yɛ nhwɛso ahorow te sɛ Stable Diffusion.
Diffusion Models (Diffusion Models)
Diffusion Models
Generative models kuw bi a wosua sɛnea wɔyɛ data denam random noise a wɔde sesa no kɔ structured outputs mu.
Nhwɛso: Stable Diffusion yɛ mfonini a ɛyɛ nokware fi nsɛm a wɔde kyerɛw no denam diffusion akwan so.
Embedding (Embedding)
Embedding
Data a wɔde kyerɛw no wɔ nɔma mu, a wɔtaa de di dwuma de kyerɛ nsɛmfua, mfonini, anaa nsɛm nkyerɛkyerɛmu.
Nhwɛso: Wɔ NLP mu no, asɛmfua 'bank' betumi anya embeddings a ɛte sɛ 'money' nanso ɛnte sɛ 'riverbank' a ɛfa nsɛm ho.
Epoch (Epoch)
Epoch
Sɛnea wɔde machine learning nhwɛso bi kyerɛkyerɛ mu no, ɛyɛ iteration a edi mũ wɔ training data set no nyinaa so.
Nhwɛso: Sɛ data set bi wɔ nhwɛso 1,000 na nhwɛso bi hu wɔn nyinaa pɛnkoro wɔ training mu a, ɛno ne epoch biako.
Ethical AI (Ethical AI)
Ethical AI
Nhyehyɛe ne dwumadi nyansapɛ a ɛma AI mfiridwuma yɛ adwuma pefee, pɛpɛɛpɛ, na ɛne ɔmanfo gyinapɛn hyia.
Nhwɛso: AI adwumayɛfo a wɔde wɔn di dwuma a ɛfa bias checks ho de siw nyiyim a ɛba wɔ nnipa kakraa bi ho no kwan.
Expert System (Expert System)
Expert System
AI nhyehyɛe bi a ɛyɛ nnipa adwumayɛfo gyinaesi tumi wɔ adwuma pɔtee bi mu denam mmara ne nsusuwii so.
Nhwɛso: Expert system a wɔde di dwuma wɔ kuayɛ mu de kyerɛkyerɛ nnɔbae mu denam asase data ne mmoawa ho nsɛm so.
Explainable AI (XAI) (Explainable AI (XAI))
Explainable AI (XAI)
AI nhyehyɛe a wɔayɛ no sɛnea ɛbɛyɛ a ne mu adwuma ne ne gyinaesi bɛyɛ nea nnipa betumi ate ase, na ɛma wɔde wɔn ho to so na wɔyɛ akontabuo.
Nhwɛso: Aduruyɛ mu AI a ɛkyerɛkyerɛ mu sɛnea ɛbɛyɛ a ɛbɛma wɔahu sɛnea ɛyɛ adwuma.
Few-shot Learning (Few-shot Learning)
Few-shot Learning
Machine learning kwan bi a wɔde kyerɛkyerɛ anaa wɔde yɛ nhwɛso bi denam nhwɛso kakraa bi a wɔde labels ayɛ so.
Nhwɛso: Sɛnea wɔde LLM yɛ mmara ho e-mail ahorow bere a wɔakyerɛkyerɛ mu nhwɛso 10 pɛ.
Fine-tuning (Fine-tuning)
Fine-tuning
Sɛnea wɔde nhwɛso a wɔakyerɛkyerɛ mu dedaw di dwuma na wɔde kyerɛkyerɛ mu wɔ data foforo, nketewa so de yɛ adwuma pɔtee bi.
Nhwɛso: Sɛnea wɔde LLM te sɛ GPT kyerɛkyerɛ mu wɔ mmara ho nkrataa mu de yɛ mmara ho boafo.
Foundation Model (Foundation Model)
Foundation Model
Nhwɛso kɛse bi a wɔakyerɛkyerɛ mu wɔ data ahorow so a wobetumi de adi dwuma wɔ adwuma ahorow pii mu.
Nhwɛso: GPT-4 ne PaLM 2 yɛ nhwɛso ahorow a ebetumi ayɛ nsɛm, nsɛmmisa ne mmuae, nkyerɛase, ne nea ɛkeka ho.
Fuzzy Logic (Fuzzy Logic)
Fuzzy Logic
Nsusuwii kwan bi a ɛfa nneɛma a ɛyɛ den ho sen sɛ ɛbɛyɛ nokware/atoro (binary) nsusuwii, a ɛboa ma wɔyɛ nsusuwii wɔ nea ɛnteɛ mu.
Nhwɛso: Wɔde di dwuma wɔ wim tebea nhyehyɛe mu de sesa ɔhyew denam fuzzy inputs te sɛ 'a bit hot' anaa 'very cold' so.
Generative Adversarial Network (GAN) (Generative Adversarial Network (GAN))
Generative Adversarial Network (GAN)
Generative model architecture bi a networks abien — generator ne discriminator — di asie de ma nea ɛba no yɛ kɛse.
Nhwɛso: GANs na wɔde yɛ deepfake video ahorow anaa wɔyɛ mfonini a ɛyɛ nokware fi sketches.
Generative AI (Generative AI)
Generative AI
Artificial intelligence kuw bi a ebetumi ayɛ nsɛm foforo — te sɛ nsɛm, mfonini, nnwom, anaa video — fi training data.
Nhwɛso: ChatGPT a ɛyɛ blog posts anaa Midjourney a ɛyɛ digital artwork fi nsɛm a wɔde kyerɛw.
Generative Pre-trained Transformer (GPT) (Generative Pre-trained Transformer (GPT))
Generative Pre-trained Transformer (GPT)
Kasa nhwɛso akɛse kuw bi a OpenAI yɛe a ɛde transformer architecture di dwuma na wɔakyerɛkyerɛ mu wɔ nsɛm pii so de yɛ kasa adwuma ahorow.
Nhwɛso: GPT-4 betumi akyerɛw essays, akyerɛkyerɛ kasa mu, na akyerɛkyerɛ nkrataa mu a ɛnyɛ den.
Genetic Algorithm (Genetic Algorithm)
Genetic Algorithm
Optimisation kwan bi a ɛfa abɔde mu nhyehyɛe ho a ɛma nsɛm kɔ anim denam mutation, crossover, ne selection so.
Nhwɛso: Wɔde di dwuma de yɛ neural network architectures a ɛyɛ adwuma yiye denam survival of the fittest a wɔde yɛ nhwɛso so.
Hallucination (Hallucination)
Hallucination
AI nhwɛso bi a ɛyɛ nsɛm a ɛyɛ nokware nanso ɛnteɛ anaa ɛnyɛ nea ɛfata.
Nhwɛso: Kasa nhwɛso bi a ɛyɛ citation a enni hɔ anaa ɛde abakɔsɛm a ɛnteɛ ma.
Heuristic (Heuristic)
Heuristic
Akwan a wɔfa so siesie nsɛm a ɛnyɛ den a ɛnyɛ nea ɛyɛ pɛpɛɛpɛ nanso ɛfata ma atirimpɔw a ɛwɔ hɔ.
Nhwɛso: Sɛnea wɔde rule of thumb di dwuma de bu akontaa sɛnea ɛbɛyɛ a wɔbɛtumi de nneɛma akɔ logistics AI nhyehyɛe mu.
Hyperparameter (Hyperparameter)
Hyperparameter
Configuration value bi a wɔahyɛ no dedaw ansa na wɔakyerɛkyerɛ machine learning nhwɛso bi, te sɛ learning rate anaa layers dodow.
Nhwɛso: Sɛnea wɔsesa batch size fi 32 kosi 128 de ma training speed ne model performance yɛ kɛse.
Inference (Inference)
Inference
Sɛnea wɔde machine learning nhwɛso a wɔakyerɛkyerɛ mu di dwuma de yɛ nsɛm anaa wɔyɛ outputs fi input data foforo.
Nhwɛso: Sɛnea wɔde fine-tuned GPT nhwɛso di dwuma de yɛ e-mail ahorow ma customer support team.
Intent Detection (Intent Detection)
Intent Detection
Adwuma bi wɔ natural language understanding mu a nhyehyɛe no hu ɔdefo no atirimpɔw anaa atirimpɔw wɔ nsɛm mu.
Nhwɛso: Wɔ chatbot mu no, sɛnea wɔhu 'Mepɛ sɛ mekyerɛw wimhyɛn' sɛ travel booking intent.
Internet of Things (IoT) (Internet of Things (IoT))
Internet of Things (IoT)
Nnwinnade a wɔde wɔn ho wɔn ho abɔ mu a wɔde sensors, software, ne mfiridwuma afoforo ahyɛ mu de boaboa data ano na wɔsesa.
Nhwɛso: Smart thermostats ne fridges a ɛkyerɛkyerɛ dwumadi data na ɛsesa settings denam AI analytics so.
Interpretability (Interpretability)
Interpretability
Sɛnea nnipa betumi ate machine learning nhwɛso bi mu adwuma ne ne gyinaesi ase.
Nhwɛso: Decision tree yɛ nea wobetumi akyerɛkyerɛ mu sen deep neural network efisɛ ne gyinaesi ahorow no yɛ nea wobetumi ahu.
Jupyter Notebook (Jupyter Notebook)
Jupyter Notebook
Open-source interactive computing environment bi a ɛma nnipa tumi kyerɛw code, hu outputs, na wɔkyerɛw analysis wɔ interface biako mu.
Nhwɛso: Data scientists de Jupyter Notebooks di dwuma de yɛ machine learning nhwɛso ahorow na wɔkyɛ nea ɛba.
K-Nearest Neighbours (KNN) (K-Nearest Neighbours (KNN))
K-Nearest Neighbours (KNN)
Simple, non-parametric machine learning algorithm bi a wÉ”de di dwuma wÉ” classification ne regression mu. ÆyÉ› gyinaesi denam training examples a É›bÉ›n no wÉ” feature space mu so.
Nhwɛso: Sɛnea wɔde KNN di dwuma de kyerɛkyerɛ aba foforo mu sɛ ɛyɛ apple anaa pear, KNN hwehwɛ aba a wɔde labels ayɛ a ɛbɛn no wɔ shape ne color mu.
Knowledge Graph (Knowledge Graph)
Knowledge Graph
Data structure bi a ɛde nodes ne edges di dwuma de kyerɛkyerɛ na ɛkora nneɛma ne wɔn nkitahodi ho nsɛm a wɔde wɔn ho wɔn ho abɔ mu.
Nhwɛso: Google's knowledge panel no yɛ adwuma denam knowledge graph a ɛde nneɛma te sɛ nnipa, mmeae, ne nsɛm abɔ mu.
Language Learning Model Optimisation (LLMO) (Language Learning Model Optimisation (LLMO))
Language Learning Model Optimisation (LLMO)
Akwan a wɔde di dwuma de ma kasa nhwɛso akɛse yɛ adwuma yiye, efficiency, anaa adaptability ma adwuma anaa nsɛm pɔtee.
Nhwɛso: Sɛnea wɔde quantisation ne instruction tuning di dwuma de ma LLM yɛ adwuma yiye ma enterprise use.
Large Language Model (LLM) (Large Language Model (LLM))
Large Language Model (LLM)
Deep learning model bi a wɔakyerɛkyerɛ mu wɔ nsɛm pii so a ebetumi ayɛ, ate ase, na abɔ nsusuwii wɔ nnipa kasa ho.
Nhwɛso: ChatGPT ne Claude yɛ LLMs a wɔakyerɛkyerɛ mu de boa wɔ nkyerɛwde, coding, ne nsɛmmisa a wɔde ma.
Latent Space (Latent Space)
Latent Space
High-dimensional abstract representation bi a ɛma inputs a ɛte sɛ nea ɛbɛn no bom, a wɔde di dwuma wɔ generative models ne embeddings.
Nhwɛso: Wɔ mfonini a wɔde yɛ nhwɛso mu no, sɛnea wɔde latent space di dwuma no betumi asesa nneɛma te sɛ brightness anaa emotion.
Learning Rate (Learning Rate)
Learning Rate
Hyperparameter a ɛho hia wɔ training mu a ɛhwɛ sɛnea model weights sesa wɔ loss gradient ho.
Nhwɛso: Learning rate a ɛkorɔn no betumi ama wɔabu minima, bere a rate a ɛba fam no ma training progress yɛ brɛoo.
Machine Learning (ML) (Machine Learning (ML))
Machine Learning (ML)
AI fã bi a ɛma nhyehyɛe tumi sua ade fi data mu na ɛma adwuma yɛ kɛse a wɔankyerɛkyerɛ mu pɔtee.
Nhwɛso: Spam filters de machine learning di dwuma de kyerɛkyerɛ e-mail ahorow mu sɛ ɛyɛ spam anaa ɛnyɛ spam denam nhwɛso ahorow so.
Model Drift (Model Drift)
Model Drift
Sɛnea model bi accuracy so tew bere kɔ so esiane data anaa environment a ɛsesa nti.
Nhwɛso: Fraud detection model bi yɛ adwuma bɔne bere a fraud tactics kɔ anim.
Model Training (Model Training)
Model Training
Sɛnea wɔde data ma machine learning model bi na wɔsesa ne parameters de tew mfomso so.
Nhwɛso: Sɛnea wɔde recommendation engine kyerɛkyerɛ mu wɔ customer purchase history so de kyerɛkyerɛ nneɛma foforo.
Multimodal AI (Multimodal AI)
Multimodal AI
AI nhyehyɛe a ebetumi ayɛ na abɔ data ahorow pii mu te sɛ nsɛm, mfonini, audio, ne video.
Nhwɛso: Model bi te sɛ GPT-4 Vision a ebetumi akenkan nsɛm na akyerɛkyerɛ mfonini mu wɔ bere koro mu.
Natural Language Processing (NLP) (Natural Language Processing (NLP))
Natural Language Processing (NLP)
AI fã bi a É›fa kÉ”mputa ne nnipa (natural) kasa nkitahodi ho. Æma mfidie tumi kenkan, te ase, na wÉ”yÉ› mmuae wÉ” nnipa kasa mu.
Nhwɛso: NLP na wɔde di dwuma wɔ voice assistants, language translation apps, ne chatbots.
Neural Network (Neural Network)
Neural Network
Machine learning model bi a ɛfa nnipa adwene ho, a ɛyɛ layers a ɛde wɔn ho wɔn ho abɔ mu (neurons).
Nhwɛso: Neural networks na ɛwɔ deep learning models a wɔde di dwuma wɔ mfonini ne kasa a wɔde hu nneɛma mu.
Noise (Noise)
Noise
Nsɛm a ɛyɛ random anaa ɛnyɛ nea ɛfata wɔ data mu a ebetumi asiw nhwɛso a ɛsom bo kwan na ɛde ɔhaw aba model performance.
Nhwɛso: Sensor mfomso anaa typo-filled data entries no betumi abu no sɛ noise.
Ontology (Ontology)
Ontology
Structured framework bi a ɛkyerɛkyerɛ na ɛkyerɛkyerɛ nkitahodi ahorow mu wɔ nsusuwii ahorow mu wɔ domain bi mu, a wɔtaa de di dwuma wɔ semantic AI nhyehyɛe mu.
Nhwɛso: Ontology bi wɔ akwahosan mu no betumi akyerɛkyerɛ sɛnea nsɛm a ɛfa nyarewa ho ne nyarewa ne nnuru ho nsɛm.
Overfitting (Overfitting)
Overfitting
Modelling mfomso bi a machine learning model bi hu noise wɔ training data mu na ɛyɛ adwuma bɔne wɔ data foforo so.
Nhwɛso: Model bi a ɛkae training answers nanso ɛntumi nni test data a wonhuu no ho dwuma no yɛ overfitted.
Predictive Analytics (Predictive Analytics)
Predictive Analytics
Sɛnea wɔde data, algorithms, ne AI di dwuma de hu nea ebetumi aba wɔ daakye mu denam abakɔsɛm data so.
Nhwɛso: Retailers de predictive analytics di dwuma de bu akontaa sɛnea nneɛma pɔtee bi ho hia.
Pre-training (Pre-training)
Pre-training
Sɛnea wɔde model bi kyerɛkyerɛ mu wɔ data kɛse, general data so ansa na wɔde ayɛ adwuma pɔtee.
Nhwɛso: GPT models no yɛ pre-trained wɔ corpora akɛse so ansa na wɔde ayɛ customer service chatbots.
Prompt Engineering (Prompt Engineering)
Prompt Engineering
Sɛnea wɔyɛ prompts a ɛyɛ adwuma yiye de kyerɛ large language models output.
Nhwɛso: Sɛnea wɔde system instructions te sɛ 'Reply as a polite tutor' ka ho no yɛ prompt engineering nhwɛso.
Quantisation (Quantisation)
Quantisation
Model compression kwan bi a ɛtew bits dodow a wɔde kyerɛ weights ne activations so, na ɛma efficiency yɛ kɛse.
Nhwɛso: Sɛnea wɔde model bi fi 32-bit kɔ 8-bit no ma performance yɛ kɛse wɔ mobile devices so.
Quantum Computing (Quantum Computing)
Quantum Computing
Computing kwan foforo bi a ɛfa quantum mechanics ho, a ɛwɔ tumi sɛ ɛma processing capabilities yɛ kɛse.
Nhwɛso: Quantum computing betumi ama AI training akɔ anim sen classical limits.
Reasoning Engine (Reasoning Engine)
Reasoning Engine
AI nhyehyɛe bi a ɛde nsusuwii a ɛfata fi nsɛm anaa data mu denam mmara anaa inference algorithms so.
Nhwɛso: AI diagnosis tool bi de reasoning engine di dwuma de hu aduruyɛ mu tebea a ebetumi aba denam nsɛm so.
Reinforcement Learning (RL) (Reinforcement Learning (RL))
Reinforcement Learning (RL)
Machine learning fã bi a agents sua ade denam wɔn environment a wɔde wɔn ho gye mu de ma cumulative rewards yɛ kɛse.
Nhwɛso: Robot bi a ɛreboa sɛnea ɛbɛnantew denam trial and error so denam RL akwan so.
Reinforcement Learning with Human Feedback (RLHF) (Reinforcement Learning with Human Feedback (RLHF))
Reinforcement Learning with Human Feedback (RLHF)
Adesua kwan bi a nnipa pɛ sɛ wɔkyerɛ AI reward signal, a wɔtaa de di dwuma wɔ fine-tuning language models.
Nhwɛso: ChatGPT no wɔakyerɛkyerɛ mu wɔ RLHF de yɛ mmuae a ɛboa na ɛyɛ ahobammɔ.
Retrieval-Augmented Generation (RAG) (Retrieval-Augmented Generation (RAG))
Retrieval-Augmented Generation (RAG)
Kwan bi a ɛde information retrieval ne generation bom, a LLM hwehwɛ nkrataa a ɛfata de ma ne mmuae yɛ kɛse.
Nhwɛso: AI boafo bi a ɛhwehwɛ na ɛkyerɛ product specs bere a ɛyɛ mmuae ma nsɛmmisa a ɛfa mfiridwuma ho.
Self-Supervised Learning (Self-Supervised Learning)
Self-Supervised Learning
Training kwan bi a model no sua nhwɛso ahorow denam ne labels a ɛyɛ fi raw data mu, na ɛtew nnipa-annotated data so.
Nhwɛso: BERT no wɔakyerɛkyerɛ mu wɔ self-supervised learning de bu akontaa nsɛmfua a ɛyɛ den wɔ nsɛm mu.
Semantic Search (Semantic Search)
Semantic Search
Search kwan bi a ɛte ɔdefo no atirimpɔw ne nsɛm nkyerɛkyerɛmu ase, ɛnyɛ keyword matching pɛ.
Nhwɛso: Sɛnea wɔhwehwɛ 'how to fix a leaking tap' no ma guides ba sɛnea ɛbɛyɛ a asɛmfua 'leaking tap' no nni hɔ wɔ document no mu.
Sentiment Analysis (Sentiment Analysis)
Sentiment Analysis
Sɛnea wɔhu adwene, adwene, anaa suban wɔ nsɛm mu, a wɔtaa kyerɛkyerɛ mu sɛ ɛyɛ positive, negative, anaa neutral.
Nhwɛso: Sɛnea wɔhwehwɛ tweets mu de hu ɔmanfo mmuae ma product foforo.
Stochastic (Stochastic)
Stochastic
Sɛnea ɛfa randomness anaa probabilistic suban ho, a wɔtaa de di dwuma wɔ generative AI ne optimisation algorithms.
Nhwɛso: GPT-4 output no sesa ma input koro no ara esiane ne stochastic decoding process nti.
Strong AI (Strong AI)
Strong AI
Wɔsan frɛ no Artificial General Intelligence (AGI), ɛfa mfidie a ɛwɔ nnipa-level cognitive capabilities wɔ domains nyinaa mu.
Nhwɛso: AI a ɛbɛba daakye a ebetumi akyerɛw novels, ayɛ nkurow ho nhyehyɛe, na asiesie ethical dilemmas pɛpɛɛpɛ.
Super Artificial Intelligence (SAI) (Super Artificial Intelligence (SAI))
Super Artificial Intelligence (SAI)
AI a ɛwɔ adwene mu a ɛboro nnipa adwene so wɔ nneɛma nyinaa mu—nsusuwii, creativity, emotional intelligence, etc.
Nhwɛso: SAI betumi ayɛ nyansahu ne nyansapɛ foforo wɔ ne ho.
Supervised Learning (Supervised Learning)
Supervised Learning
Machine learning kwan bi a wɔde kyerɛkyerɛ models wɔ labelled data so de sua input-output mappings.
Nhwɛso: Sɛnea wɔkyerɛkyerɛ model bi sɛ ɛbɛkyerɛkyerɛ e-mail ahorow mu sɛ ɛyɛ spam anaa ɛnyɛ spam denam abakɔsɛm nhwɛso ahorow so.
Synthetic Data (Synthetic Data)
Synthetic Data
Artificially generated data a ɛyɛ real-world data, a wɔtaa de di dwuma wɔ training mu bere a real data yɛ kakraa bi anaa ɛyɛ sensitive.
Nhwɛso: Sɛnea wɔyɛ synthetic medical images de kyerɛkyerɛ diagnostic models a wonni patient privacy.
Token (Token)
Token
Nsɛm bi a LLMs yɛ — a ɛyɛ asɛmfua anaa asɛmfua fã.
Nhwɛso: Asɛm 'Hello world!' no wɔakyekyɛ mu tokens 3: 'Hello', 'world', ne '!'.
Tokenisation (Tokenisation)
Tokenisation
Sɛnea wɔkyekyɛ nsɛm mu kɔ tokens mu de ma model bi yɛ adwuma.
Nhwɛso: Wɔ NLP mu no, 'ChatGPT is great' no yɛ ['Chat', 'G', 'PT', 'is', 'great'].
Transfer Learning (Transfer Learning)
Transfer Learning
Sɛnea wɔde nimdeɛ fi adwuma biako mu di dwuma de ma adesua yɛ kɛse wɔ adwuma foforo a ɛfa ho, na ɛtew training time ne data ho hia so.
Nhwɛso: Sɛnea wɔde model bi a wɔakyerɛkyerɛ mu wɔ English text so di dwuma de yɛ sentiment analysis wɔ kasa foforo mu.
Transformer (Transformer)
Transformer
Neural network architecture bi a ɛde attention mechanisms di dwuma de yɛ sequential data, a wɔde di dwuma kɛse wɔ LLMs.
Nhwɛso: BERT, GPT, ne T5 yɛ transformer-based models.
Underfitting (Underfitting)
Underfitting
Bere a model bi yɛ simple dodo sɛ ɛbɛtumi ahu nhwɛso ahorow wɔ training data mu, na ɛde performance bɔne ba.
Nhwɛso: Linear model bi a ɛbɔ mmɔden sɛ ɛbɛka complex image classifications ho asɛm no betumi ayɛ underfit.
Unsupervised Learning (Unsupervised Learning)
Unsupervised Learning
Adesua kwan bi a models hu nhwɛso anaa clusters wɔ unlabelled data mu.
Nhwɛso: Sɛnea wɔde nnipa bom denam purchasing behaviour so a wonni predefined labels.
User Intent (User Intent)
User Intent
Atirimpɔw anaa atirimpɔw a ɛwɔ ɔdefo no nsɛmmisa anaa nkitahodi akyi.
Nhwɛso: Ɔdefo bi a ɔkyerɛw 'how to bake a cake' no pɛ sɛ ɔhwehwɛ recipe.
Validation Set (Validation Set)
Validation Set
Data fã bi a wɔde di dwuma de hwehwɛ model performance mu wɔ training mu na wɔde tune hyperparameters.
Nhwɛso: Wɔde di dwuma de hu overfitting ansa na wɔayɛ testing a etwa to.
Vector Database (Vector Database)
Vector Database
Database bi a wɔayɛ no sɛnea ɛbɛyɛ a ɛbɛkora na ahwehwɛ vector embeddings a wɔde di dwuma wɔ AI adwuma ahorow te sɛ similarity search ne RAG.
Nhwɛso: Pinecone ne Weaviate yɛ vector databases a wɔde kora text anaa image embeddings.
Vector Embedding (Vector Embedding)
Vector Embedding
Data a wɔde kyerɛw no wɔ nɔma mu a ɛkora semantic meaning ne nkitahodi ahorow wɔ vector space mu.
Nhwɛso: Nsɛmfua 'king' ne 'queen' no wɔ embeddings a ɛte sɛ nea ɛwɔ gender differences.
Virtual Assistant (Virtual Assistant)
Virtual Assistant
AI-powered software agent bi a ɛboa nnipa ma wɔyɛ adwuma denam nkitahodi anaa voice commands so.
Nhwɛso: Siri, Alexa, ne Google Assistant yɛ virtual assistants a agye din.
Voice Recognition (Voice Recognition)
Voice Recognition
Mfiridwuma a ɛkyerɛkyerɛ na ɛsesa kasa kɔ nsɛm anaa adwuma mu.
Nhwɛso: Voice typing ne voice commands no gyina voice recognition systems so.
Weak AI (Weak AI)
Weak AI
AI nhyehyɛe a wɔayɛ no sɛnea ɛbɛyɛ a ɛbɛyɛ adwuma pɔtee bi a wonni general intelligence.
Nhwɛso: Chess-playing AI bi a ɛntumi nte kasa ase anaa ɛntumi nni kar no yɛ weak AI nhwɛso.
Web Scraping (Web Scraping)
Web Scraping
Sɛnea wɔde wɔn ho yɛ information extraction fi websites, a wɔtaa de di dwuma de boaboa training data anaa wɔhwɛ content.
Nhwɛso: Scraping real estate listings de kyerɛkyerɛ property valuation model.
Weight (Weight)
Weight
Parameter bi wɔ neural networks mu a ɛkyerɛ sɛnea node biako nya ɔfoforo so.
Nhwɛso: Weights sesa wɔ training mu de tew model no mfomso so.
Whisper (Whisper)
Whisper
Speech-to-text model bi a OpenAI yɛe a ebetumi akyerɛw audio wɔ kasa ahorow pii mu.
Nhwɛso: Whisper betumi akyerɛw lectures ne podcasts a ɛyɛ pɛpɛɛpɛ.
YAML (YAML)
YAML
Human-readable format bi a wɔde yɛ data serialisation, a wɔtaa de di dwuma wɔ configuration files wɔ machine learning workflows.
Nhwɛso: Sɛnea wɔkyerɛkyerɛ model parameters wɔ YAML file mu de yɛ training wɔ PyTorch.
Zero-shot Learning (Zero-shot Learning)
Zero-shot Learning
Model bi tumi sɛ ɛbɛyɛ adwuma a wɔankyerɛkyerɛ mu pɔtee denam general knowledge a wɔde di dwuma so.
Nhwɛso: Model bi a ɛma mmara ho nsɛmmisa mmuae a wɔankyerɛkyerɛ mu pɔtee wɔ mmara ho data so.
Zettabyte (Zettabyte)
Zettabyte
Digital data biako a ɛyɛ sextillion (10^21) bytes, a wɔtaa de di dwuma de kyerɛ internet data kɛse.
Nhwɛso: Global internet traffic no boroo 1 zettabyte afe biara wɔ afe 2016 mu.
