Semester.ly

Johns Hopkins University | EN.540.605

Modern Data Analysis and Machine Learning for Chembes

3.0

credits

Average Course Rating

(4.07)

This class will provide an introduction for chemical and biomolecular engineering students to modern methods of measuring and testing hypotheses using experimental or computational data. The course will cover methods of regression and data analysis such as linear and nonlinear regression, Bayesian analysis and principal or independent component analysis. The course will introduce concepts of machine learning including linear and nonlinear separation, neural networks, Gaussian processes and will provide exposure to deep learning concepts. The course will focus generally on image data and will consider topics of image processing, feature extraction and will cover for general data dimensionality reduction. Familiarity with computer programming (ideally Python), statistics and linear algebra are prerequisites.

Spring 2013

(4.36)

Spring 2014

(3.93)

Spring 2023

(3.93)

Spring 2013

Professor: Rebecca Schulman

(4.36)

The best aspects of this course included the access to cutting-edge research published in recent journal articles, the intriguing lectures, and the instructor who was very enthusiastic and approachable. Some students felt the quizzes were excessively difficult to prepare for. One suggestion was to integrate real data with some of the predictive algorithms. Another suggestion included a second pair of eyes for the instructor, who often handed out homework with a lot of typos and errors. Prospective students should know it is highly advantageous to have some exposure to MATLAB and basic ODEs before taking this course.

Spring 2014

Professor: Rebecca Schulman

(3.93)

The highlights of this course included the fascinating topics covered, rigorous assignments, and a final project rather than traditional final exam. Many students appreciated the format of assignments and found they learned more from reading and understanding research papers verses cramming for an exam. The most negative aspects of the class included the disorganization of some lectures, the disconnection from real world applicability, and the lack of any discussion of the literature. Suggested improvements to the course included more discussion in class, enforcing the prerequisite of MATLAB, and a better chalkboard in the room. Prospective students should be familiar and comfortable with MATLAB and study lectures in preparation for quizzes. This class is interesting and highly recommended. 68

Spring 2023

Professor: Rebecca Schulman

(3.93)

Lecture Sections

(01)

No location info
R. Schulman
16:30 - 17:45