Раздел 1 · Основы

ИИ, машинное и глубокое обучение

ИИ — широкая область. ML обучается закономерностям данных, глубокое обучение использует многослойные нейросети. Продукт — это большая система вокруг модели.

Начальный20 минут

После раздела вы сможете

Distinguish AI, machine learning, deep learning, a model and a complete system.
Recognize supervised, self-supervised and reinforcement learning.
Choose a baseline before reaching for a large model.

Vue interactive

Explorez cette leçon à votre rythme

Choisissez une vue. Le détail n’apparaît que lorsque vous le demandez.

Interaction locale · aucune donnée envoyée

Concept 1 / 5

AI is a field, not one technology

AI includes rules, search, planning, optimization, probabilistic methods, machine learning and hybrids. The simplest adequate method is often the most reliable.

Частые заблуждения

Afficher +

Calling every automation “AI” hides whether the system follows rules, searches, predicts or learns.
A benchmark score does not guarantee performance on your users, language, data or failure modes.
AGI and ASI are debated capability goals, not proven properties of current general-purpose models.

Ключевые слова

Afficher +

AI
algorithm
model
data
parameter
training
loss
gradient
generalization
overfitting

Первичные и официальные источники

Voir les sources +

Deep Learning bookOpen reference by Goodfellow, Bengio and Courville on machine learning and deep learning foundations.NIST AI Risk Management FrameworkAuthoritative framework for governing, mapping, measuring and managing AI risks.

ИИ, машинное и глубокое обучение

Начальный20 минут

После раздела вы сможете

Distinguish AI, machine learning, deep learning, a model and a complete system.
Recognize supervised, self-supervised and reinforcement learning.
Choose a baseline before reaching for a large model.

Vue interactive

Explorez cette leçon à votre rythme

Choisissez une vue. Le détail n’apparaît que lorsque vous le demandez.

Interaction locale · aucune donnée envoyée

Concept 1 / 5

AI is a field, not one technology

AI includes rules, search, planning, optimization, probabilistic methods, machine learning and hybrids. The simplest adequate method is often the most reliable.

Частые заблуждения

Afficher +

Calling every automation “AI” hides whether the system follows rules, searches, predicts or learns.
A benchmark score does not guarantee performance on your users, language, data or failure modes.
AGI and ASI are debated capability goals, not proven properties of current general-purpose models.

Ключевые слова

Afficher +

AI
algorithm
model
data
parameter
training
loss
gradient
generalization
overfitting

Первичные и официальные источники

Voir les sources +