Rugved Mhatre

MS. Computer Engineering @NYU Ex-Oracle

1 minute read

“Machine Learning” course (ECE-GY 6143) at NYU Tandon by Anna Choromanska

nyu-header

The “Machine Learning” course (ECE-GY 6143) at NYU Tandon by Anna Choromanska offers a comprehensive journey through fundamental and advanced concepts in machine learning, blending theoretical knowledge with practical applications to address real-world challenges.

Course Overview

The course covers topics such as regression, empirical risk minimization, cross-validation, additive models, and logistic regression, progressing to perceptron algorithms, stochastic gradient descent, backpropagation, and neural networks. Subsequent weeks delve into generalization guarantees, VC-dimension, support vector machines, kernels, and probability models. Students explore discrete and continuous probability models, Gaussian distributions, principal component analysis (PCA), Bayesian inference, and clustering methods such as K-Means and Expectation-Maximization. Later topics include advanced probabilistic models, graphical models, the Bayes Ball and Junction Tree algorithms, and Hidden Markov Models (HMM). The course features a midterm in Week 7 and a final exam in Week 14, with weekly assignments reinforcing concepts. Designed for students with a background in linear algebra, calculus, probability, and programming (Python preferred), this course equips participants with the skills to develop machine learning models and apply them effectively across various domains.

Course Work and Grading

Note: This is subject to change

  • Homeworks - 30%
  • Midterm Exam - 30%
  • Final Exam - 40%
  • Participation in Seminars (Extra) - 10%

Review

This Machine Learning course serves as a solid introduction to the field, especially for those looking to build a strong theoretical foundation. It provides an in-depth mathematical understanding of why machine learning models work the way they do, which is essential for truly grasping the concepts behind ML. The course is well-structured, covering a wide range of topics from regression and classification to probability models, clustering, and graphical models, making it ideal for anyone wanting to understand the principles behind machine learning algorithms.

However, the course leans heavily toward theory, with less emphasis on implementing models or gaining extensive hands-on experience. For those interested in a more practical, implementation-focused approach, the course by Prof. Fraida Fund might be a better choice. While this course is invaluable for understanding the mathematical backbone of ML, those looking to immediately apply models in real-world scenarios might find Prof. Fund’s course more aligned with their goals. That said, this course is an excellent starting point for anyone serious about mastering machine learning concepts.

You May Also Enjoy