Below is a list of externally-hosted workshops and seminars to support your data science literacy and analytics capacity.
To request event promotion on our website, please contact dash.coordinator@ubc.ca.
Past Events
2021
October 5: Biomedical Imaging and Artificial Intelligence (BMIAI) Invited Lecture Series - Reading Race: AI Recognises Patient's Racial Identity in Medical Images
TUESDAY, OCTOBER 5, 2021
12:00PM
Join UBC's Biomedical Imaging & AI and Data Science & Health Research Clusters in welcoming Dr. Judy Gichoya for a virtual invited lecture.
ABOUT THE SPEAKER
Dr. Judy Gichoya is a multidisciplinary researcher, trained as both an informatician and a clinically active radiologist. She is an assistant professor at Emory university, and works in Interventional Radiology and Informatics. She is seconded to the National Institutes of Health as a data scholar to help with the Open Data Science Platform (OSDP) component of the DSI Africa Initiative to “Harness Data Science for Health In Africa”. Her career focus is on validating machine learning models for health in real clinical settings, exploring explainability, fairness, and a specific focus on how algorithms fail. She has worked on the curation of datasets for the SIIM (Society for Imaging Informatics in Medicine) hackathon and ML committee. She volunteers on the ACR and RSNA machine learning committees to support the AI ecosystem to advance development and use of AI in medicine.
October 14: Biomedical Imaging and Artificial Intelligence (BMIAI) Fall Research Showcase
THURSDAY, OCTOBER 14, 2021
9:00AM - 11:00PM
Join BMIAI this Fall to discover the exciting research being conducted by their multi-disciplinary members.
The Research Showcase aims to expose participants to BMIAI's research activities and provide a platform for engagement and collaboration.
The Showcase will also feature the launch of BMIAI's new project repository. The project repository will be an online database of research questions where members, collaborators, and potential industry partners can submit their project ideas. The objective of the repository is to increase collaborations and act as a matchmaking mechanism. If you would like to submit a research problem to the repository, you can do so here.
The day will include:
-
Lightning Talks by BMIAI members
-
Virtual video presentations exhibiting BMIAI members' work
-
Official launch of the BMIAI repository
2022
January 20: AI in Med Lecture - Dr. Anne Martel
WEDNESDAY, JANUARY 12, 2022
12:00 – 1:00 PM
ZOOM MEETING
ABSTRACT
The introduction of scanners that are capable of digitizing microscopic slides at high magnification has led to an explosion of interest in computational pathology in general and deep learning applied to whole slide images (WSIs) in particular. In my lab at Sunnybrook, we are developing AI models that can detect cancer, automatically segment regions of interest, and learn predictive and prognostic models that can be used to guide treatment decisions. In this talk I will outline some of the unique challenges of working with these extremely large WSIs and discuss some of the approaches that we have developed to overcome the problems of sparse annotations and weak, noisy labels, including self-supervision and multiple instance learning. I will also outline some of the challenges in deploying AI algorithms to the clinic.
SPEAKER
Anne Martel is a Professor in Medical Biophysics at the University of Toronto, a Senior Scientist at Sunnybrook Research Institute, and a Vector Faculty Affiliate. Her research program is focused on medical image and digital pathology analysis, particularly on applications of machine learning for segmentation, diagnosis, and prediction/prognosis. In 2006 she co-founded Pathcore, a software company developing complete workflow solutions for digital pathology. Dr Martel is an active member of the medical image analysis community and is a fellow of the MICCAI Society which represents engineers and computer scientists working in this field. She has served as board member of MICCAI and is currently on the editorial board of Medical Image Analysis.
January 28: Data Privacy Day 2022 Discussion
FRIDAY, JANUARY 28, 2022
11:00 AM – 12:00 PM
ZOOM MEETING
Did you know that recent changes to British Columbia's Freedom of Information and Protection of Privacy Act could greatly affect how you learn and work at UBC? Come join the Privacy Matters @ UBC team as to celebrate Data Privacy Day on Jan 28 with a very important discussion about changes to Data Residency Restrictions in British Columbia.The event will discuss: How will the removal of certain data residency restriction affect UBC from a security, financial, and procurement perspective? How will these changes affect the use of cloud-based technology and digital tools for faculty, staff, and students? All UBC faculty and staff are encouraged to register and join the conversation.
February 23: Ethics in Technology
WEDNESDAY, FEBRUARY 23, 2022
12:00 – 1:15 PM
ZOOM MEETING
Ethics in Technology is the second installation of Ethics for UBC, a five-part speaker series that will explore the current landscape of ethics scholarship and education across the Vancouver and Okanagan campuses of our university.
In attending this interactive 75-min panel discussion, you will have the opportunity to learn about the current ethics-related endeavours taking place at UBC, pressing ethical issues that exist across a wide range of disciplines, and the ways in which you can become more involved.
This panel discussion will focus on ethics across a variety of disciplines that involve technology, from engineering and architecture to neuroscience and nursing.
ABOUT THE SPEAKERS
Welcome by: Gail Murphy, PhD; Professor, Computer Science; Vice President – Research and Innovation, University of British Columbia
Moderated by: Ari Rotenberg, BA; MSc Candidate, Experimental Medicine; Neuroethics Canada, University of British Columbia
Panelists:
Shahria Alam, PhD; Professor, School of Engineering; Director, Green Construction Research Training Centre, University of British Columbia – Okanagan Campus
Fionn Byrne, MLA; Assistant Professor, School of Architecture and Landscape Architecture, University of British Columbia
Julie Robillard, PhD; Assistant Professor of Neurology, Faculty of Medicine; Faculty, Neuroethics Canada, University of British Columbia
Patricia Rodney, PhD; Associate Professor Emerita, School of Nursing, University of British Columbia
June 27-30: Bootcamp on Systems-Data Science - Practical Combining Data Science and Systems Science for Health
JUNE 27-30, 2022
ONLINE & IN-PERSON OPTIONS
The Bootcamp on Systems-Data Science presents industrial-strength methods combining the power of machine learning, big data and dynamic modeling to improve health decision making. In addition to more than a dozen previous health uses, these methods have been at the heart of regular reporting by the instructor to inform health system decision-making in SK (daily), for each Canadian province (via PHAC), and for First Nations reserves across 9 Canadian provinces (via FNIHB).
This bootcamp characterizes these techniques, how to use them effectively, provides overviews of how they work, describes their use with other approaches, case studies & example implementations. The event offers a particular focus on how these methods can be used with “big data” offering high volume, velocity, variety, and veracity, with examples showing the particular ways that they can regularly reground models with data from wastewater sampling, social media, smartphones/wearables, and search data in addition to traditional epidemiological data.
In greater detail, systems science and data science are two rapidly developing areas of computational science that have been demonstrated to offer tremendous capacity for informing health understanding, and which are applied by a growing number of projects in health and health care. While each of these approaches taps the power of computational models, they have traditionally largely been pursued in isolation from each other. Such fragmentation is particularly unfortunate, because the techniques are not merely highly compatible -- for example, in each using computational or informatics mechanisms to provide temporally and locationally fine grained longitudinal understanding across multiple generative pathways -- but synergistic, with each tradition opening strong opportunities for empowering the other, and with the combination of both yielding opportunities for insight and improved decision making far beyond the sum of what each can bring in isolation. We present here a proven set of Systems-Data Science methods that achieve this “whole greater than the sum of its parts” and that allow each approach to better fulfill its full potential.
August 22-27: Agent-Based & Hybrid Modeling Bootcamp & Incubator for Health & Health Care - Conceptualization, Formulation, Implementation for Insight
AUGUST 22-27, 2022
HYBRID (ONLINE & IN-PERSON OPTIONS FOR BOOTCAMP; IN-PERSON ATTENDANCE REQUIRED FOR INCUBATOR)
The Agent-Based and Hybrid Modeling Bootcamp & Incubator for Health Researchers is an intensive, hands-on tutorial and dynamic model incubator that seeks to provide systematic, practical and technically detailed guidance in conceptualizing, designing and implementing agent-based and hybrid system science modeling for health and health care, using familiar language and concepts. This year’s bootcamp & incubator will focus on agent-based modeling, but with a particular emphasis on insightful, practical and performant hybrids with System Dynamics modeling and discrete event modeling, such as for health care service delivery and patient flow.
To help ease participants in overcoming the learning curve in applying the techniques taught herein to domains of individual interest, those registering for the “incubator” side of the event will further leverage the extensive experience of the instructor, other expert modelers, and teaching assistants (TAs), who will provide ongoing advice, guidance, tips and close hands-on assistance as these participants and their partnered TAs work to build, explore, and debug the start of their own models addressing participant research interests. Participants interested in taking their initial models far further can elect to remain for the full-scale model incubator/hackathon the subsequent week, in which the instructor and teams of TAs will partner with attendees to deliver more evolved models capable of contributing substantive health insights. Attendance to the incubator portion of this event is constrained by the availability of teaching assistants -- please reserve early to ensure an incubator spot.
Using lectures and guided exercises, this bootcamp will introduce a broad set of topics in designing & building agent-based and hybrid modeling, offer optional material for those seeking additional depth of coverage, and emphasize the complementary insights when agent-based and hybrid modeling is combined with traditional health sciences approaches. Course material is drawn from the instructor’s semester-long interdisciplinary courses at MIT and the University of Saskatchewan, as well as previous bootcamps, including those hosted at UCLA, University of New South Wales, NCSU/UNC, Flinders University, Deakin University, the Sax Institute, and beyond.
Participants will be provided with over 50 example models drawn from across diverse spheres of health and health care. Health & health care modeling case studies for chronic disease, communicable illnesses, health service delivery, and environmentally driven infectious diseases from around the world will support participants in helping to conceptualize fruitful model designs, scopes and the form of effective modeling projects. Such models will showcase Agent-Based, System Dynamics and Discrete Event modeling, including -- but not limited to -- use of agents some of whose states are driven by System Dynamics stocks & flows, agent-based epidemiological models linked to discrete event models capturing patient care pathways and patient flow, highly performant hybrid models in which System Dynamics is used for low-risk populations and in which agents are individuated at a certain point in risk progression or where interacting agents drive flow rates, multi-scale models, and case studies combining all three types of modeling.
The optional final day of the bootcamp will give participants the opportunity to brainstorm with the instructor and teaching assistants concerning ideas for modeling projects, and to discuss concrete design and implementation strategies for such projects.
August 29 - September 2: Agent-Based & Hybrid Modeling for Health & Health Care Hackathon
AUGUST 29 - SEPTEMBER 2, 2022
IN-PERSON
This hackathon will provide participant wrap-around assistance in conceptualizing, building, exploring, and debugging models addressing their particular research or practitioner interests and vision. To do so, this event will leverage the extensive experience of the instructor and teams of teaching assistants, who will provide ongoing advice, guidance, tips and hands-on assistance to attendees, allowing delivery of more evolved models capable of contributing substantive health insights. Attendees for this event can either attend this hackathon as a stand-alone event or stay on from the bootcamp & incubator the previous week (thereby building on their learning from that event and continuing on any model substantively started in the incubator portion of that event).
Attendance to the hackathon is constrained by the availability of TA teams -- please reserve early to ensure a spot.
August 29: CAIDA Seminar - Differentially Private Fine-Tuning of Language Models
AUGUST 29, 2022
11:00AM - 12:00PM
UBC VANCOUVER CAMPUS, ICCS X836
ABSTRACT
We give simpler, sparser, and faster algorithms for differentially private fine-tuning of large-scale pre-trained language models, which achieve the state-of-the-art privacy versus utility tradeoffs on many standard NLP tasks. We propose a meta-framework for this problem, inspired by the recent success of highly parameter-efficient methods for fine-tuning. Our experiments show that differentially private adaptations of these approaches outperform previous private algorithms in three important dimensions: utility, privacy, and the computational and memory cost of private training. On many commonly studied datasets, the utility of private models approaches that of non-private models. For example, on the MNLI dataset we achieve an accuracy of 87.8% using RoBERTa-Large and 83.5% using RoBERTa-Base with a privacy budget of ϵ=6.7. In comparison, absent privacy constraints, RoBERTa-Large achieves an accuracy of 90.2%. Our findings are similar for natural language generation tasks. Privately fine-tuning with DART, GPT-2-Small, GPT-2-Medium, GPT-2-Large, and GPT-2-XL achieve BLEU scores of 38.5, 42.0, 43.1, and 43.8 respectively (privacy budget of ϵ=6.8,δ= 1e-5) whereas the non-private baseline is 48.1. All our experiments suggest that larger models are better suited for private fine-tuning: while they are well known to achieve superior accuracy non-privately, we find that they also better maintain their accuracy when privacy is introduced. No knowledge of differential privacy will be assumed. Based on joint work with Da Yu, Saurabh Naik, Arturs Backurs, Sivakanth Gopi, Huseyin A. Inan, Janardhan Kulkarni, Yin Tat Lee, Andre Manoel, Lukas Wutschitz, Sergey Yekhanin, and Huishuai Zhang. Paper appeared in ICLR 2022, and available on arXiv (https://arxiv.org/abs/2110.06500).
ABOUT THE SPEAKER
Dr. Gautam Kamath is an Assistant Professor at the David R. Cheriton School of Computer Science at the University of Waterloo. He has a B.S. in Computer Science and Electrical and Computer Engineering from Cornell University, and an M.S. and Ph.D. in Computer Science from the Massachusetts Institute of Technology, where he was advised by Constantinos Daskalakis. His research interests lie in methods for statistics and machine learning, with a focus on challenges which arise in modern data analysis, including data privacy and robustness. He was a Microsoft Research Fellow, as a part of the SimonsBerkeley Research Fellowship Program at the Simons Institute for the Theory of Computing. He is recipient of an NSERC Discovery Accelerator Supplement, and was awarded the Best Student Presentation Award at the ACM Symposium on Theory of Computing in 2012.
September 22-23: World Data Congress 2022 "AI & Data Science - What's Next"
AUGUST 23-26, 2022
SAN FRANCISCO, USA
The WDC conference series is where professionals meet the best and brightest innovators in AI and Data Science. The conference aims to bring together top industry and C-level executives to help you understand how AI and data science can transform your business, leading to the next industrial revolution. Two days Conference is part of the World Data Congress series, which is one of the leading series dedicated to Data Science & AI worldwide.
World Data Congress aims to deliver the advancements of AI & data science, its management, innovation, application, latest research, and comprises of keynotes sessions, expert presentations, high-stake panels and round-table discussions to highlight AI & Data Science scenario, industries automation process, AI implementation and research and other relevant topics. WDC-2022 will provide a valuable opportunity for innovators, researchers and industry experts to exchange their ideas.
Industries covered include Finance, IT, Healthcare, Biotech, Pharma, Travel, Energy, Manufacturing, Retail, Marketing, Transportation, and more.