• A
  • A
  • A
  • АБB
  • АБB
  • АБB
  • А
  • А
  • А
  • А
  • А
Обычная версия сайта
2023/2024

Машинное обучение II

Лучший по критерию «Полезность курса для расширения кругозора и разностороннего развития»
Лучший по критерию «Новизна полученных знаний»
Статус: Маго-лего
Когда читается: 1 модуль
Охват аудитории: для всех кампусов НИУ ВШЭ
Преподаватели: Мартинез Саито Марио
Язык: английский
Кредиты: 3
Контактные часы: 30

Course Syllabus

Abstract

Prerequisites: Basic knowledge of Statistics and Python.This course aims to provide state-of-the-art techniques of mathematical statistics as well as new and modern methods of machine learning. What is more important, this course will focus on practical activities and it will allow students to learn on their mistakes. Moreover, skills and knowledge obtained during this course could be applied to almost any field of science and industry. Students will know statements of all major machine learning problems and mathematical details of the most important data analysis methods and algorithms. Also, they will obtain skills, such selection of an appropriate method for solving particular data analysis problems, performance of basic data processing and visual analysis, features generation for subsequent machine learning, application machine learning libraries, algorithm's selection hyperparameters, critically evaluate the obtained results and redesign data-processing pipelines, ability to solve real-world data science problems using modern machine learning techniques.
Learning Objectives

Learning Objectives

  • Students will get familiar with the basic concepts and methods of classical machine learning, and will acquire working knowledge of how to select and apply machine learning techniques to solve real world problems.
Expected Learning Outcomes

Expected Learning Outcomes

  • Getting acquainted with the methods of model-free reinforcement learning, such as value learning.
  • Recognizing and applying the most popular types of decision trees.
  • Understanding how backpropagation works in neural networks with multiple layers.
  • Deliver a presentation on a topic related to machine learning
  • Understanding the basics of model-based reinforcement learning and its application to complex problems
  • Applying machine learning methods to practical tasks such as data and text processing
  • Learning how to configure neural nextworks for multiclass recognition and how to interpret patterns on high-dimensional spaces through dimensionality reduction techniques
  • Students can prepare data for analysis
  • Students can make tables with summary statistics, static and dynamic graphs
  • Students can make regression analysis
  • Students can apply basic classification algorithms
  • Students can apply ensemble models for classification or regression tasks
  • Students can apply basic techniques of dimensionality reduction
  • Students can make data clustering
Course Contents

Course Contents

  • Data preprocessing
  • EDA. Data visualization
  • Regression
  • Classification
  • Ensembles
  • Dimensionality reduction
  • Cluster analysis
Assessment Elements

Assessment Elements

  • non-blocking History, latest advances, development frameworks, coding project
  • non-blocking Exam
  • non-blocking Participation
    Active participation during the seminars
Interim Assessment

Interim Assessment

  • 2023/2024 1st module
    0.4 * Exam + 0.3 * History, latest advances, development frameworks, coding project + 0.3 * Participation
Bibliography

Bibliography

Recommended Core Bibliography

  • Pattern recognition and machine learning, Bishop, C. M., 2006

Recommended Additional Bibliography

  • Hastie, T., Tibshirani, R., & Friedman, J. H. (2009). The Elements of Statistical Learning : Data Mining, Inference, and Prediction (Vol. Second edition, corrected 7th printing). New York: Springer. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=277008

Authors

  • Martinez Saito MARIO
  • SHALOM KSENIYA VLADIMIROVNA