Duke +DataScience Recorded Experiences Repository

Duke +DS is providing free and easy access to recordings of all its recent virtual learning sessions, including slides and coding instructions, via its Recorded Experiences Repository. The +DS team launched the learning repository in the fall of 2020 to support learner flexibility and asynchronous learning. The resource has since been accessible to the entire Duke community including students, faculty, and staff.

In 2020, +DS held 49 learning experiences, beginning with in-person learning experiences and pivoting to virtual learning in spring 2020 amid the COVID-19 pandemic. Over the summer of 2020, +DS hosted an 8-week series on data science methods with direct application to the COVID-19 pandemic; the series achieved 1,509 attendances across 12 events.

The +DS team received numerous requests for recorded content from the attendees of its COVID+DS series held between June and August, which led to the creation of the +DS Recorded Experience Repository.

“Many Duke students who sign up for +DS sessions are now living overseas due to the pandemic,” said +DS Program Coordinator Tiffany Torres. “The repository provides them access to valuable content, which they can view at their convenience.”

The Fall 2020 Virtual Learning Experiences were the first to be archived in the learning repository, followed by the January 2021 “AI for Everyone” sessions. These experiences were part of the Duke Winter Breakaway, which attracted 372 graduate, undergraduate, and professional students.

The +DS Recorded Experiences Repository covers a wide range of topics from learning sessions held over the past year, including natural language processing, image analysis, applied case studies, product development, and PyTorch, plus fundamental topics including concepts in machine learning and ethical issues with emerging AI.

Some highlights of recordings available on the +DS learning repository are:

  • Slides and recorded sessions from The Duke Winter Breakaway “AI for Everyone” session, including the extended Q&A
  • Content from Fall 2020 Virtual Learning Experiences
  • Archived recordings from the 2019 lunch-and-learns (a series of lectures that may be of particular interest to the medical community)
  • Participant instructions for PyTorch coding sessions

Below is a comprehensive list of recordings archived in the +DS Recorded Experiences Repository:

Duke+DataScience logo
Recorded Experiences Repository

Fall 2020 +DS Virtual Learning Opportunities

  • Introduction to PyTorch for Deep Learning, Serge Assaad
  • Introduction to Basic Concepts in Machine Learning, Ricardo Henao
  • The Transformer Network for Natural Language Processing, Lawrence Carin
  • Deep Learning with PyTorch for Natural Language Processing, Liqun Chen
  • Convolutional Neural Networks for Image Analysis, Tim Dunn
  • Deep Learning with PyTorch for Image Analysis, Rachel Draelos
  • bespokeDS vLE: Effective Data Visualization, Matthew Hirschey, and Cédric Scherer
  • COVID-19 and the Telehealth Transformation: Insights into MyChart using Natural Language Processing, Jedrek Wosik and Shijing Si
  • Analysis of CT Scan Imaging Data with Machine Learning: Classification, Detection, and Segmentation of Abnormalities, Rachel Draelos
  • Natural Language Processing with LSTM Recurrent Neural Networks, Lawrence Carin
  • Deep Learning and ‘Smart’ Microscopy, Roarke Horstmeyer
  • Squash, Gardener and the future of AI in political fact-checking, Bill Adair and Jun Yang
  • Generative Adversarial Networks, Ricardo Henao
  • How to build successful machine learning products (that solve real problems and make money!), Jon Reifschneider
  • On the role of data science in FinTech; Jimmie Lenz, in conversation with Larry Carin
  • Overview of Ethical Issues with Emerging AI, Nita Farahany
  • Deep Learning from the Perspective of the Experimental Biologist Part 1 of 4: Introduction, Akshay Bareja
  • Deep Learning from the Perspective of the Experimental Biologist Part 2 of 4: What is a neural network? Akshay Bareja
  • Deep Learning from the Perspective of the Experimental Biologist Part 3 of 4: Using a multilayer perceptron to classify blood cell images, Akshay Bareja
  • Deep Learning from the Perspective of the Experimental Biologist Part 4 of 4: Using a convolutional neural network to classify blood cell images, Akshay Bareja

 Winter Breakaway Sessions:

  • What is a “machine” and how does one do “learning” on a computer?, David Carlson
  • Logistic regression and a multi-layered perceptron (MLP); introduction to deep learning, Ricardo Henao
  • Analysis of electronic health records using a MLP, Rachel Draelos
  • Use of CNNs for image analysis, Tim Dunn
  • Recurrent neural networks for analysis of sequential data, Larry Carin
  • CNNs for analysis of pathology images, David Dov
  • Analysis of neuroscience data with recurrent neural networks, David Carlson
  • Analysis of American politicians via natural language processing analysis of their tweets, Sunshine Hillygus and Greg Spell
  • Introduction to PyTorch for machine learning, Serge Assaad
  • The transformer network for natural language processing, Larry Carin

Recorded content from 2019 Lunch-and-Learns:

  • Digital Pathology: Identifying Thyroid Malignancy with Deep Learning,
    Danielle Range, Yoni Cohen, Ricardo Henao
  • A Window to the Brain: Analysis of Retinal Images with Deep Neural Networks,
    Sharon Fekrat, Felipe Medeiros, Dilraj Singh Grewal, Lawrence Carin
  • Early Autism Screening with Machine Learning, Geraldine Dawson and Guillermo Sapiro
  • Recommending MyChart Responses with Natural Language Processing, Jedrek Wosik and Ricardo Henao