Scribd
Upload
55 views

Lecture1 AML

This document provides information about an advanced machine learning course. It outlines the instructor, contact information, teaching assistants, required skills, course structure including assignments, projects, exams and a schedule of lecture topics.

Uploaded by

happy_user
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
55 views

Lecture1 AML

This document provides information about an advanced machine learning course. It outlines the instructor, contact information, teaching assistants, required skills, course structure including assignments, projects, exams and a schedule of lecture topics.

Uploaded by

happy_user
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Advanced Machine Learning

Lecture 1

Anna Choromanska
[email protected]
https://engineering.nyu.edu/faculty/anna-choromanska
Department of Electrical and Computer Engineering
New York University Tandon School of Engineering
Information about the class

Instructor: Anna Choromanska


Contact: [email protected]
Office hours: Monday, 9.15 am (zoom)
Information about the class

Instructor: Anna Choromanska


Contact: [email protected]
Office hours: Monday, 9.15 am (zoom)

CA:
Kristi Topollai
Contact: [email protected]
Office hours: Mondays 11am-1.30pm and Thursdays 11am-1.30pm
Office hours location: Room 825, 370 Jay Street
Grader:
Amrutha Patil
Contact: [email protected]
Information about the class

Put on calendars:
Tutorial on the usage of NYU HPC: available online or on sessions
offered by NYU HPC
ECE Seminar Series in Modern AI (mandatory):
Jamie Shotton: 27th of February (in person)
Arthur Gretton: 11th of March (in person)
Drew Bagnell: 15th of April (in person)
Samantha Wood: 6th of May (in person)
Information about the class
Necessary skills:
Mathematical maturity:
https://en.wikipedia.org/wiki/Mathematical_maturity
Information about the class
Necessary skills:
Mathematical maturity:
https://en.wikipedia.org/wiki/Mathematical_maturity
Course Work and Grading:
Homeworks (3 in total) - 20% - deadline is at midnight
Project - 25% - deadline is at midnight
Midterm - 20%
Final - 20%
Seminar Series Participation (must participate in all seminars)- 15%
Information about the class
Necessary skills:
Mathematical maturity:
https://en.wikipedia.org/wiki/Mathematical_maturity
Course Work and Grading:
Homeworks (3 in total) - 20% - deadline is at midnight
Project - 25% - deadline is at midnight
Midterm - 20%
Final - 20%
Seminar Series Participation (must participate in all seminars)- 15%
Academic Honesty Policy:
http://engineering.nyu.edu/life/student-affairs/
community-standards-procedures
http://www.nyu.edu/about/policies-guidelines-compliance/
policies-and-guidelines/
academic-integrity-for-students-at-nyu.html
Information about the class

Homework write-ups should be preferably written in Latex.


Project write-ups should be preferably written in NeurIPS conference
template.
Information about the class

Homework write-ups should be preferably written in Latex.


Project write-ups should be preferably written in NeurIPS conference
template.
Computing environments (recommended, but students are allowed to choose
other environments):
Matlab - all students should have access to Matlab
PyTorch (for deep learning) - Linux is required
Install from: http://pytorch.org/
Vowpal Wabbit (VW; for online learning algorithms) - Linux is required
Install from: https://github.com/JohnLangford/vowpal_wabbit
Information about the class

Homework write-ups should be preferably written in Latex.


Project write-ups should be preferably written in NeurIPS conference
template.
Computing environments (recommended, but students are allowed to choose
other environments):
Matlab - all students should have access to Matlab
PyTorch (for deep learning) - Linux is required
Install from: http://pytorch.org/
Vowpal Wabbit (VW; for online learning algorithms) - Linux is required
Install from: https://github.com/JohnLangford/vowpal_wabbit
Computational resources:
NYU HPC (Torch and VW) - account permission is required
Information about the class

Project rules:
Each group consists of two students. Larger groups are not allowed.
Information about the class

Project rules:
Each group consists of two students. Larger groups are not allowed.
Those who plan projects related to their research or even plan publishing
should indicate that (PhDs and Masters with excellent academic record
and possibly publication record).
Information about the class

Project rules:
Each group consists of two students. Larger groups are not allowed.
Those who plan projects related to their research or even plan publishing
should indicate that (PhDs and Masters with excellent academic record
and possibly publication record).
Students can propose their-own topics too.
Information about the class

Lecture schedule:

Class 1 (01/23/2024) and Class 2 (01/30/2024): Perceptron, Exponentiated


Gradient, Experts advice, online multi-class classification
Class 1: Project topics are released.
Homework 1 is released and due 02/20/2024
Class 3 (02/6/2024) and Class 4 (02/13/2024): Deep learning, feed-forward
neural networks, back-propagation algorithm, convolutional neural networks (CNNs),
regularization, adversarial networks, auto-encoders, LISTA, transformers,
non-convexity and optimization landscape in deep learning
Class 3: Due date for selecting projects.
Class 5 (02/20/2024): Empirical Risk Minimization, PAC bounds
Due date for Homework 1.
Homework 2 is released and due 03/26/2024
Class 6 (02/27/2024): Seminar with Jamie Shotton.
Information about the class

Class 7 (03/05/2024): Occam’s Razor, VC dimension, VC inequality, Structural


Risk Minimization, Support Vector Machines (SVM), kernels
Class 8 (03/12/2024): MIDTERM
Class 9 (03/26/2024) and 10 (04/02/2024): Introduction to optimization
(convexity, Lipschitzness and smoothness, Jensen’s inequality, convergence rates),
optimization methods (GD, SGD, Netwon, BFGS, LBFGS, ISTA, FISTA, bound
majorization), convergence guarantees
Class 9: Due date for Homework 2.
Class 9: Homework 3 is released and due 04/16/2024
Information about the class

Class 11 (04/09/2024): Expectation Maximization (EM): basics and modern


statistical view
Class 12 (04/16/2024): Curse of dimensionality, Johnson–Lindenstrauss lemma,
principal component analysis (PCA), clustering (k-center clustering, k-means
clustering, hierarchical clustering, doubling algorithm, cover tree algorithm, spectral
clustering)
Due date for Homework 3.
Class 13 (04/23/2024) and Class 14 (04/30/2024):
Project presentations: 15-20 minutes per presentation including
demo video and a clear statement of what was done by each
student.
Class 13: Due date for Projects.
Class 15 (05/07/2024): FINAL EXAM

You might also like