Learn

+DS provides training modules and learning experiences grounded in generalizable data science content, while partnering with individual units or groups to develop additional specialized content.

Online Learning

Coursera Online Modules

The online +DS modules introduce the basics of data science, across multiple important application domains. These online modules are used as prerequisites for the in-person learning experiences listed below. Together, the online content supports the in-person "flipped" learning experiences. 

Module 1: Introduction to Machine Learning

Logistic regression, a simple machine learning (ML) method, is introduced and then extended to the multilayered perceptron (MLP), a fundamental neural network. Conceptual understanding is provided to motivate the form of the MLP, and its use.

Module 2: Basics of Model Learning

Concepts in learning model parameters are introduced, as well as model validation and testing. Learning based on stochastic gradient descent is addressed, allowing scaling to large datasets ("big data").

Module 3: Image Analysis and the Convolutional Neural Network

The convolutional neural network (CNN) is developed for image analysis, including details of the model and its underlying components. Model training is covered, as well as transfer learning and fine-tuning.

Module 4: Introduction to Natural Language Processing

Application of neural networks to natural language processing (NLP) is covered, from simple neural models to the more complex. The fundamental concept of word embeddings is discussed, as well as how such methods are employed within model learning and usage for several NLP applications.

Please note that Coursera for Duke is accessible to only Duke students, faculty, and staff. If you are not a member of the Duke community, you can access the public version of this Coursera course: https://www.coursera.org/duke.

Recorded Content from Machine Learning Summer School

In addition to the aforementioned Coursera content, +DS offers recordings of Duke’s Machine Learning Summer School (MLSS), which was help in June of 2018. If you are a Duke student, staff or faculty member, you can review these classroom recordings on Panopto, with accompanying slides and links to github code demos.

In-Person Learning Experience

In addition to the online content, +DS offers in-person opportunities to dive deeper into the information introduced in the online modules. These learning experiences will be developed to target diverse units at Duke: from those that desire a broad understanding of what is possible with data science, and those who wish to use data-science tools (software) without a need for deep understanding of underlying methodology, to those who desire a rigorous technical proficiency of the details and methodology of data science.

See past learning experiences

Upcoming In-Person Learning Experiences
Tuesday, September 3 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: David Carlson

The basic concepts of neural networks are introduced, with a focus on intuition. The simpler and widely used logistic regression model is introduced first, and from this the neural network is introduced as a generalization.

Thursday, September 5 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Timothy Dunn

The convolutional neural network (CNN) represents the current state-of-the-art for image and video analysis, and is increasingly used for analyzing time series and other data with spatial or sequential structure.

Monday, September 9 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Lawrence Carin

Natural language processing (NLP) is a field focused on developing automated methods for analyzing text, and also for computer-driven text generation (synthesis, for example in translation). Neural networks have recently become the state-of-the-art method for NLP.

Wednesday, September 11 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Kevin Liang

TensorFlow is Google’s open-source framework for developing, learning and training neural networks, and it is widely employed.

Tuesday, September 17 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Lawrence Carin

Reinforcement learning is a branch of machine learning, in which in algorithm learns a good policy for acting in an environment of interest, based on experience.

Wednesday, September 18 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Matthew Kenney

With recent advancements in natural language generation, AI models can now produce highly coherent generative text. In this IPLE, we will focus on creative and practical applications for language generation.

Thursday, September 19 - 4:00pm to 6:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Ricardo Henao Giraldo

Generative adversarial networks (GANs) are a new tool in machine learning, that leverage advances in deep neural networks. Using GANs, one can develop a computer model that is capable of synthesizing highly realistic images, such as human faces and interesting art.

Tuesday, September 24 - 4:00pm to 5:00pm
Location: Bostock Library, 1st floor, Edge Workshop
Instructor: Nita Farahany

Artificial intelligence (AI) can reduce costs, improve efficiency, and potentially improve accuracy in many critical areas of life that impact humans. And yet, many of the tools of AI lack transparency, have inherent biases, and are difficult to govern.