Machine Learning Winter School

Machine Learning Winter School

See information about the Machine Learning Summer School being held in June 2019.

January 6-8, 2019 (this event has already happened)

Together Duke is pleased to announce the Machine Learning Winter School (MLWS), being offered for the first time in January 2019, as a three-day class that provides lectures on the fundamentals of machine learning, and modern deep learning.


Machine learning is a field characterized by development of algorithms that are implemented in software and run on a machine (e.g., computer, mobile device, etc.). Each such algorithm is characterized by a set of parameters, and particular parameter settings yield associated algorithm characteristics. The algorithms have the capacity to learn, based on observed data. By “learn” it is meant that the algorithm can rigorously quantify which parameter settings are best matched to the data of interest. After algorithm parameters are so learned, the associated model ideally captures the underlying characteristics of the data. The algorithm, with learned parameters, may subsequently be applied to new data, with the goal of making predictions or learning insights. Machine learning methodology is primarily concerned with designing appropriate models/algorithms for datasets and problems of interest, plus the capacity to learn the model parameters given data (with challenges manifested when that data is of a massive scale).

In the context of prediction, one may be interested in developing algorithms that are capable of automatically interpreting data in a healthcare setting, to improve clinical care. In this case, the healthcare data may be radiological images, doctor notes, and/or a history of patient care (e.g., previous diagnoses, medications taken, etc.). In healthcare, the goal is to use machine learning to make improved diagnoses and recommendations for care. Similar concepts are of interest in business, where one may be interested in tailoring advertising and products to individuals. In education, machine learning may be used to tailor educational material to the level and interests of each student. Machine learning is increasingly making an impact in almost all areas of personal and professional life.

Recently, with increasing access to massive datasets, and to significant advances in computing resources, the quality of machine learning performance (e.g., prediction accuracy) has improved markedly. Further, over the last five years, significant advances have been made in a subfield of machine learning called “deep learning.”

This class will focus on the areas of machine learning that have made the biggest advances in utility over the last several years, particularly deep learning. The class will concentrate on methods that allow machine-learning algorithms to train effectively on massive datasets, i.e., “big data.” Emphasis will be placed on the latest methods for image and video analysis, as well as natural language processing. In addition to these application areas, the course is meant to provide participants with a strong foundation in deep learning fundamentals.

Professor Lawrence Carin, of the Duke Electrical & Engineering Department, will lead the MLWS, and several other Duke professors will also lecture.

Who Should Attend

The Machine Learning Winter School (MLWS) is targeted to individuals interested in learning about machine learning, with a focus on recent algorithms, like deep learning. The MLWS will introduce the mathematics and statistics at the foundation of modern machine learning. Additionally, the MLWS will provide details on case studies, demonstrating how this technology is used in practice.

The MLWS is meant to provide value to students at multiple levels of mathematical sophistication (including with limited such background). On each day of the MLWS, an initial emphasis will be placed on presenting the concepts in as intuitive a manner as possible, with minimum math. As the concepts are developed further, more math will be introduced, but only the minimum necessary to explain the concepts. Finally, case studies will show how the technology is used in practice, and these discussions should be accessible to most students (concepts emphasized over detailed math). Strength in mathematics and statistics is a significant plus, and will make all MLWS material accessible; however, it is not required to benefit from much of the program.

Program Format

The three-day class will provide lectures on the fundamentals of machine learning, and modern deep learning. During each day of the MLWS, the morning sessions will be devoted to introducing fundamentals on the area of focus that day, and the afternoon sessions will be devoted to case studies, of the methods applied to specific application areas. While software details will not be discussed formally in class, participants will be pointed to available code such that the methods may be applied.

Each day of the MLWS will run from 9:00am-4:00pm and will be arranged as follows:

  • 9:00-10:15am    Lecture 1: Mathematically-light introduction to the focus of the day
  • 10:45am-noon  Lecture 2: Mathematically rigorous discussion of the focus of the day

In the afternoon there will be two sessions running in parallel:

In one room we will have:

  • 1:30-2:30pm     Case Study 1: Example of the machine learning concept in practice
  • 3:00-4:00pm     Case Study 2: Example of the machine learning concept in practice

and in another room during the same time periods there will be an introduction-to-TensorFlow session (students can choose to attend the Case Studies or TensorFlow discussions).

The detailed daily schedules are as follows:

January 6:
Morning lectures on introduction to neural nets and basics of machine learning: David Carlson
Case Study 1: Multi-layered perceptrons for predicting at-risk patients: Rachel Draelos
Case Study 2: Face recognition and privacy methods: Guillermo Sapiro
TensorFlow Session: TensorFlow basics and multi-layered perceptrons with TensorFlow: Kevin Liang

January 7:
Morning lectures on convolutional neural networks and image analysis: Tim Dunn
Case Study 1: Image segmentation in medicine: Matt Engelhard
Case Study 2: Object recognition in images for security: Kevin Liang
TensorFlow Session: TensorFlow applied to image analysis: Johnny Sigman

January 8:
Morning lectures on natural language processing (NLP): Larry Carin
Case Study 1: Sentiment analysis with NLP: Larry Carin
Case Study 2: Question-answer NLP with neural networks: Larry Carin
TensorFlow Session: TensorFlow applied to NLP with neural networks: Liqun Chen

Program Details: Location, Registration and Cost

MLWS is being held in Schiciano Auditorium, which is in the Fitzpatrick Center for Interdisciplinary Engineering, Medicine and Applied Sciences FCIEMAS on Duke’s West Campus. Visitor parking is available in the nearby Bryan Center Parking Garage.

Students (with a valid ID, at Duke or other universities) will pay a course fee of $100; the fee for non-students is $500, payable through the registration site. All fees are non-refundable.

MLWS will be available for up to 200 participants. We will maintain a waitlist beyond the maximum registration, and will contact those on the waitlist as spots become available.


Lecturers in MLWS include:

Lawrence Carin, Ph.D.

David Carlson, Ph.D.

Rachel Draelos, M.D./Ph.D. student

Tim Dunn, Ph.D.

Matthew Engelhard, M.D., Ph.D.

Kevin Liang, PhD. student

Guillermo Sapiro, Ph.D.

John Sigman, Ph.D.

MLWS is a new program offering from

Registration for the January 2019 Machine Learning Winter School closes on January 4, 2019 at 11:59pm. For help or for more information, contact Carolyn Mackman at