Certificat de perfectionnement professionnel en intelligence artificielle appliqu茅e

脌 propos de ce programme

Ce programme pratique et avanc茅 de perfectionnement professionnel non cr茅dit茅, offert en ligne et en classe, est con莽u pour donner aux participants des connaissances et des comp茅tences concr猫tes et pertinentes pour l鈥檌ndustrie, afin de faire d鈥檈ux des experts en science des donn茅es et des d茅veloppeurs en intelligence artificielle (IA) performants. Le programme vise 脿 d茅velopper des comp茅tences d鈥櫭﹙aluation, de conception, d鈥櫭﹍aboration et d鈥檃m茅lioration d鈥檃lgorithmes d鈥橧A par l鈥檈ntremise de projets pratiques et de r茅solution de probl猫mes. Les participants 茅laboreront un portfolio de projets en IA. Le programme est offert en anglais et doit 锚tre termin茅 en deux ans.

Ce programme est offert en anglais et doit 锚tre compl茅t茅 dans deux ans.

Crit猫res d鈥檃dmission

Certificat de perfectionnement professionnel en science des donn茅es et apprentissage automatique de l鈥櫭塩ole d鈥櫭ヾucation permanente de l鈥橴niversit茅 McGill;

A) Dipl么me de baccalaur茅at (ou sup茅rieur) ou formation 茅quivalente dans l鈥檜n des domaines suivants : g茅nie, informatique, g茅nie logiciel, math茅matiques, statistiques ou discipline connexe.

B) Et l'exp茅rience ant茅rieure en apprentissage automatique et solides comp茅tences de programmation dans Python.

C) Et un test de comp茅tences en ligne sur l鈥檃pprentissage automatique et sur Python OU聽 cours YCBS 255 Apprentissage statistique automatique.

Note :

Les 茅tudiants d鈥櫭e adulte (21 ans et plus) qui ne r茅pondent pas aux crit猫res ci-dessus, mais qui ont une exp茅rience significative et pertinente en d茅veloppement de logiciels, en apprentissage automatique ou en science des donn茅es et qui ont d茅j脿 r茅ussi des cours pertinents en calcul, en statistiques ou en apprentissage automatique, verront leurs demandes 茅valu茅es au cas par cas.

Qui devrait participer

Ce programme est surtout pertinent pour les professionnels qui ont d茅j脿 suivi le Certificat de perfectionnement professionnel en science des donn茅es et apprentissage automatique de l鈥櫭塩ole d鈥櫭ヾucation permanente de l鈥橴niversit茅 McGill et qui d茅sirent pousser plus loin leurs comp茅tences en apprentissage automatique. Il convient 茅galement aux d茅veloppeurs de logiciels ayant de l鈥檈xp茅rience en science des donn茅es ou en apprentissage automatique ainsi que de solides comp茅tences de programmation dans Python.

Objectifs

脌 la fin du programme, vous pourrez :

茅valuer et choisir l鈥檃lgorithme ou le mod猫le appropri茅 pour r茅soudre un probl猫me li茅脿 l鈥橧A;
analyser les erreurs et concevoir des strat茅gies pour les r茅soudre;
int茅grer les r茅centes avanc茅es en apprentissage automatique et en intelligence artificielle dans le processus d鈥檃nalyse des donn茅es;
mettre en pratique des strat茅gies et des m茅thodes avanc茅es pour optimiser les mod猫les d鈥檃pprentissage automatique;
concevoir et cr茅er des syst猫mes logiciels int茅grant des mod猫les d鈥檃pprentissage automatique et des algorithmes d鈥檌ntelligence artificielle.

Droits de scolarit茅

Vous trouverez toute l鈥檌nformation sur les droits de scolarit茅 et les autres frais pour ce programme, notamment pour les formules 脿 temps partiel et 脿 temps plein, 脿 l鈥檃dresse mcgill.ca/continuingstudies/fr/certificats-de-perfectionnement-professionnel-frais-de-cours.

Aide financi猫re et bourses

Comme les programmes de certificat de perfectionnement professionnel ne donnent pas droit 脿 des unit茅s, ils ne sont pas admissibles 脿 l鈥檃ide financi猫re ni aux bourses comme ceux du programme de pr锚ts et bourses (Aide financi猫re aux 茅tudes) du Qu茅bec.

Veuillez aussi noter que des re莽us fiscaux officiels (formulaire聽T2202A et relev茅聽8) peuvent 锚tre fournis pour les droits de scolarit茅 associ茅s aux cours de ces programmes.

Cours

Non-Credit Courses

2 Cours requis (12 UEC)

YCNG 228 Predictive & Classification Modelling (6 CEUs)

Predictive modelling is a process that uses data and statistics to predict outcomes using data models. These models are used to detect fraud, optimize marketing campaigns, reduce risk, manage resources and improve operations. This course will familiarize students with how to design experiments and optimize ML models and interpret their output. Students will test hypotheses by using different techniques, design experiments, use machine learning algorithms to evaluate the output of different machine learning models, and explore methods to increase precision and/or accuracy.

YCNG 229 Neural Networks & Deep Learning (6 CEUs)

Artificial neural networks are a set of algorithms, inspired by the way human brain processes information, which are designed to recognize patterns. Deep learning is one of machine learning methods based on artificial neural networks. Neural networks and deep learning offer the most powerful techniques to deal with different aspects of data science such as natural language processing, computer vision and time series analysis. Building on the knowledge and skills acquired during YCBS 258 Practical Machine Learning course, this course will focus on the practical application of neural network models with the objective to develop students鈥� ability to implement them using Python and Keras.

3 Cours compl茅mentaires (17-18 UEC)

YCNG 230 Intelligent Agents & Reinforcement Learning (6 CEUs)

Intelligent agents are programs that can be applied to autonomously solve real-world optimization and planning problems, as well as help deal with incomplete information or uncertain environments. Reinforcement Learning is an approach to build goal-oriented intelligent agents, which help find strategies to optimize a desired outcome. In this course, students will explore the development of intelligent agents using different techniques, algorithms and approaches; the design and implementation of systems that exhibit intelligent behaviour through an end-to-end project; practical application of the most current programming tools, search methods, knowledge representation using logic and probability, planning and decision making under uncertainty and constraint satisfaction problems; reinforcement learning fundamentals including design of intelligent agents and integration with deep learning.

YCNG 231 Deep Learning for Computer Vision (6 CEUs)

Computer vision is one of the fields of study, which has benefited tremendously from the latest advancements in artificial intelligence. Computer vision is concerned with helping machines learn to 鈥渟ee鈥� and understand the context of their environment. It involves acquiring, processing, transforming, modifying, analyzing and understanding digital images or videos. The result is the extraction of data needed to understand the content of digital images, infer something from the environment, thus allowing complex decisions to be made. In this course, participants will learn more about computer vision from acquisition of image data to complex decision-making using deep learning methods and techniques. Image acquisition through sensors; feature detection; image classification, detection, segmentation. Convolutional Neural Networks (CNN) architectures and their applications. Most current computer vision applications, libraries and image databases will be explored.

YCNG 232 Natural Language Processing Fundamentals (6 CEUs)

Natural Language Processing (NLP) is a subfield of artificial intelligence concerned with the ability of computers to process, understand and interpret human languages. Deep learning has revolutionized NLP by improving many aspects of the field since its transition from classic linguistic approaches. This course will provide an overview of Natural Language Processing fundamentals such as language models, word embeddings, Recurrent Neural Networks (RNNs), wide variety of applications of neural networks in NLP, intrinsic and extrinsic evaluation, most current NLP tools.

YCNG 233 Time Series Analysis Fundamentals (6 CEUs)

Time series is an area of machine learning concerned with the analysis of series of data points ordered in time. Time series analysis is used in many fields to predict trends: from industrial machinery data, smart home, precision agriculture, cyber security, customer usage/engagement, marketing, asset management, finance, etc. Due to the temporal aspect of the data, time series data require specific preprocessing, feature engineering, algorithms, and validation strategy. This course will dive into the fundamentals of applying artificial intelligence techniques on time series data, statistical, machine and deep learning models for time series. Emphasis is placed on the main applications of time series analysis: forecasting, clustering and anomaly detection. Data science techniques applicable to time series. Deep learning methods, rolling predictions, online learning, backtesting.

YCNG 234 Internet of Things (5 CEUs)

The Internet of Things (IoT) is an important source of relevant data for different industries. It permits to collect and process information about the physical world in real time. This course will examine applications of IoT, the fundamental technologies that support IoT devices and the transmission, processing, the analysis of the data they generate, and the fundamental software and hardware technologies that support IoT devices; data transmission protocols; tools to process the data generated by IoT devices; data analysis techniques that enable common applications related to IoT devices.

YCNG 235 Recommender Systems (6 CEUs)

Have you ever wondered how the message: 鈥淐ustomers like you also purchased the following items鈥︹€� appears on your screen when you are shopping online? Recommender Systems use machine learning algorithms that help users discover new products and services. There exists a wide spectrum of recommender systems applications that help users choose movies, restaurants, music to listen to, jobs to apply to, products to purchase, social media profiles, among many others. This course will explore fundamental concepts and techniques in recommender systems: similarity models, non-personalized, content-based, and hybrid systems; association rules mining; collaborative filtering: user-, item-, and graph-based models; matrix factorization; graph recommenders, sequential recommenders, evaluation of recommender systems.

Contacts

Career and Professional Development
(Non-credit courses and programs)

info.conted@mcgill.ca

Phone: +1 514-398-5454

680 Sherbrooke Street West, Suite 1029 Montreal, Quebec, Canada H3A 2M7

Horaires d'ouverture

Monday to Friday
9:00 a.m. to 5:00 p.m.

香蕉视频

Main navigation