Workshops

Technical Workshops


Past Workshops

When: Tuesday, February 25, 2025 - 3:30 p.m.-6:00 p.m.

Location: Bruininks Hall - Room 412 

Registration: https://z.umn.edu/22525DSMMATechWorkshop

Materials:https://github.com/SigProcessing/ml-reproducibility-workshop     

Workshop Facilitator: Waheed U. Bajwa, Rutgers University-New Brunswick  

Description: Computational reproducibility, defined as the ability to consistently recreate the results of a computational analysis using the same data, code, and methods, is a cornerstone of reliable and impactful research. In this 2.5-hour workshop, participants will explore the principles of computational reproducibility and learn why it is essential for avoiding setbacks in their own projects, accelerating research progress, and enhancing recognition in their field. The workshop will begin with a general discussion on the definition, importance, and challenges of computational reproducibility, addressing common pitfalls such as unmanaged sources of randomness, inconsistent environments, and undocumented workflows. Participants will then engage in hands-on activities to gain practical experience with tools and techniques that make reproducibility achievable. Topics will include best practices for managing randomness in computational experiments, structuring projects for reproducibility, using Git and GitHub for version control, managing dependencies with virtual environments, containerizing workflows with Docker, and assigning Digital Object Identifiers (DOIs) to make research outputs citable and publicly accessible. Although the examples will focus on machine learning and use Python as the primary language, the concepts and strategies discussed are broadly applicable to other data-intensive fields and extend to languages such as MATLAB, R, and Julia. Whether you are a machine learning researcher or a data scientist working on complex analyses, this workshop will provide the skills and insights needed to make your work reproducible, robust, and impactful.

When: Monday, March 25, 2024 - 4:00 p.m.-5:30 p.m.

Location: Tate 201-20

Workshop Facilitator: Xuan Bi, PhD, Carlson School of Management

Description: In recent years, there has been a growing demand to develop

efficient recommender systems which track users’ preferences and

recommend potential items of interest to users. In this workshop, we will

navigate through the basic elements of recommender systems, its

principles, possible applications, recent development, and ongoing

challenges. Furthermore, we will have some hands-on experience to build 

our own recommender systems, make recommendations, and evaluate 

their performance on real-world datasets.

Materials

When: Monday, April 3, 2023 - 2:30 p.m.-4:30 p.m.

Presenter: Muhammed Saleem Cholayil

Description: In this workshop, we will introduce DeepClean - a 

Convolutional Neural Network architecture that is designed for removing

environmental and instrumental noise couplings from Gravitational Wave 

data. To understand the nature of the noise couplings, DeepClean uses

the data from additional sensors that track the motion of the environmental 

and instrumental sources that are responsible for the noise contamination. 

The workshop will start with a presentation that will give a brief introduction 

to noise regression problems using ML and cover the details of the 

DeepClean architecture as well as how to train it for the best possible 

results. In the hands-on session, we will take a short segment of LIGO raw 

data and aim to subtract some specific noise couplings using DeepClean. 

Materials

Learning using Zooniverse Citizen-Science Data

When: Friday, December 2, 2022 - 2:00 p.m.- 4:00 p.m.

Workshop Facilitators: Ramana Kumar Sankar,

Kameswara Bharadwaj Mantha

Description: Every scientific domain has seen an explosion in the volume 

of data, necessitating the development of robust pipelines for data

reduction, and the application of automated machine/deep learning 

frameworks. Mining these datasets, performing quantitative analysis, and

designing machine learning frameworks to synthesize useful data products 

are critical skills for aspiring data scientists in the era of big data analytics. 

In this workshop, we introduce tabular and image-based data products 

from the Zooniverse citizen-science platform and explore ways of reducing 

these data. We will also introduce the concepts of deep learning and will

walk you through an implementation and training of a Convolutional Neural 

Network-based classifier on the data mined from an example 

Zooniverse dataset. Through this workshop, you will learn the basics of 

data acquisition and analysis in Python, along with data processing

techniques for developing machine models using the PyTorch deep 

learning framework.

Materials

Zoom Recording


Monday, February 28, 2022 - 3:30 p.m.-5:30 p.m. 

November 15-19, 2021 (Virtual) 

September 29-30, 2021 (Online)

    August 16-20, 2021 (Virtual)

    Tuesday, January 26, 2021 - 4:00 p.m. - 6:00 p.m.

    November 19-20, 2020

    Tuesday, November 10, 2020 - 10:00 a.m.

Professional Workshops


Past Workshops

When: October 15, 2024 from 3:30 p.m.-5:30 p.m. 

Location: Room 130 of Physics and Nanotechnology Building 

Workshop Facilitator Bio: Mariann Johnson is a wellbeing and mindfulness instructor for the University of Minnesota’s Earl E. Bakken Center for Spirituality and Healing. She has studied and practiced mindfulness meditation for over 25 years and is a certified Mindfulness-Based Stress Reduction instructor through Brown University’s Mindfulness Center. Before dedicating her professional life to teaching mindfulness, Mariann served as an organization development consultant and mediator, working with leaders of Fortune 500 companies, government agencies and nonprofit organizations. Her writings on mindful leadership have appeared in the Huffington Post and Mindful Magazine.

    Description:This experiential session will offer information and   

    inspiration to enhance personal and team wellbeing. The University of  

    Minnesota’s Earl E. Bakken Center for Spirituality & Healing’s wellbeing 

    model will be highlighted and practical, everyday strategies for mitigating 

    the effects of personal and team stress will be discussed. Mindful self-

    leadership will be explored as an approach for strengthening one’s 

    personal agency, and mindfulness will be presented as both an innate 

    capacity and an evidence-based resource for strengthening wellbeing and 

    emotional resilience. Participants will be invited to participate in 

    mindfulness practices, including mindful communication. 


When: Tuesday, February 27, 2024, 4:00 p.m.-6:00 p.m.

Location: Tate 201-20 Seminar Room

Workshop Facilitator: Brian Sostek, Teaching Specialist, UMN

Description:  Whether speaking to a conference room of 500, giving a chalk talk to a hiring committee, explaining your research in front of a poster, or participating in a three-minute thesis competition, presenting should ideally feel like leading a friend you like along a path you know well to a destination that you want to share with them. In this introduction to the fundamental elements of science communication, we’ll focus on understanding audience experience (AX), and designing presentations that are enticing, engaging, enjoyable, and effective for the people who you want to reach. 


When: Monday, January 22, 2024, 4:00 p.m.-5:00 p.m.

Location: Keller Hall 3-180

Workshop Facilitator: Kelly Nolan, an attorney-turned-time management strategist

Description: Struggling to find time for classes, your capstone research project, interviews, clubs, coffee chats, a social life, and make time for yourself? You're not the only one. Join us for a workshop to learn actionable time management strategies to help you manage it all with less stress and more breathing space. This workshop will be led by Kelly Nolan, an attorney-turned-time management strategist. After experiencing overwhelm as a young patent litigator in Boston, Kelly figured out a time management system to help her show up in the ways that she wanted to at work and at home – without requiring her brain to somehow magically remember it all. She now empowers others to manage their personal, family, and career roles with less stress and more calm clarity using realistic time management strategies. Her system, the Bright Method, has been featured in Bloomberg Businessweek, and her work has been published in Forbes, Fast Company, and Parents. Learn more at kellynolan.com. Join us, and walk away with a new level of clarity and empowerment about how to manage everything on your plate.

Materials

November 13, 2023, 4:30 p.m.-5:30 p.m.

Materials 

Zoom Recording 

March 1, 2023, 6:00 p.m.-700 p.m.

February 20, 2023 - 2:30 p.m.-4:30 p.m.

November 15, 2022 - 4:00 p.m.-5:30 p.m.

October 11, 2022 - 4:00 p.m.-5:30 p.m.

April 11, 2022 -  3:00 p.m.-5:00 p.m. 

March 14, 2022 - 3:30 p.m.-5:30 p.m.

November 6, 2021, 10:00 a.m.-1:00 p.m.

    March 16, 2021, 4:00 p.m.-5:30 p.m.

September 28,2021 - 2:00-3:30 p.m.

Presented by: Dr. Kumi Smith (Epidemiology and Community Health)


Women in Data Science Symposium

Past Workshops