Select a calendar:
Filter November Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for November
-
Shifting the Frame: The Labors of ImageNet and AI Data
Wed, Nov 06, 2024 @ 02:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr. Alex Hanna, Director of Research, Distributed AI Research Institute (DAIR)
Talk Title: Shifting the Frame: The Labors of ImageNet and AI Data
Abstract: Artificial intelligence (AI) technologies like ChatGPT, Stable Diffusion, and LaMDA have led a multi-billion dollar industry in generative AI, and a potentially much larger industry in AI more generally. However, these technologies would not exist were it not for the immense amount of data mined to make them run, low-paid and exploited annotation labor required for labeling and content moderation, and questionable arrangements around consent to use these data. Although datasets used to train and evaluate commercial models are often obscured from view under the shroud of trade secrecy, we can learn a great deal about these systems by interrogating certain publicly available datasets which are considered foundational in academic AI research.
In this talk, I investigate a single dataset, ImageNet. It is not an understatement to say that without ImageNet, we may not have the current wave of deep learning techniques which power nearly all modern AI technologies. I begin from three vantage points: the histories of ImageNet from the perspective of its curators and its linguistic predecessor WordNet, the testimony of the data annotators which labeled millions of ImageNet images, and the data subjects and the creators of the images within ImageNet. Academically, I situate this analysis within a larger theory and practice of infrastructure studies. Practically, I point to a vision for technology which is not based on practices of unrestricted data mining, exploited labor, and the use of images without meaningful consent.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: Dr. Alex Hanna is Director of Research at the Distributed AI Research Institute (DAIR). A sociologist by training, her work centers on the data used in new computational technologies, and the ways in which these data exacerbate racial, gender, and class inequality. She also works in the area of social movements, focusing on the dynamics of anti-racist campus protest in the US and Canada. She holds a BS in Computer Science and Mathematics and a BA in Sociology from Purdue University, and an MS and a PhD in Sociology from the University of Wisconsin-Madison.
Dr. Hanna has published widely in top-tier venues across the social sciences, including the journals Mobilization, American Behavioral Scientist, and Big Data & Society, and top-tier computer science conferences such as CSCW, FAccT, and NeurIPS. Dr. Hanna serves as a Senior Fellow at the Center for Applied Transgender Studies, and sits on the advisory board for the Human Rights Data Analysis Group and the Scholars Council for the UCLA Center for Critical Internet Inquiry.
She is a recipient of the Wisconsin Alumni Association’s Forward Award, has been included on FastCompany’s Queer 50 and Go Magazine’s Women We Love lists, and has been featured in the Cal Academy of Sciences New Science exhibit, which highlights queer and trans scientists of color.
With Emily M. Bender, Dr. Hanna runs the Mystery AI Hype Theater 3000 series, playfully and wickedly tearing apart AI hype for a live audience online on Twitch and on their podcast.
Host: CAIS
More Info: https://cais.usc.edu/events/usc-cais-webinar-with-dr-alex-hanna/
Location: Michelson Center for Convergent Bioscience (MCB) - 101
Audiences: Everyone Is Invited
Contact: Thomas Lord Department of Computer Science
Event Link: https://cais.usc.edu/events/usc-cais-webinar-with-dr-alex-hanna/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
-
Algorithmic Tools for Redistricting: Fairness via Analytics
Wed, Nov 20, 2024 @ 02:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr. David Shmoys, Laibe/Acheson Professor and Director of the Center for Data Science for Enterprise & Society - Cornell University
Talk Title: Algorithmic Tools for Redistricting: Fairness via Analytics
Abstract: The American winner-take-all congressional district system empowers politicians to engineer electoral outcomes by manipulating district boundaries. To date, most computational solutions focus on drawing unbiased maps by ignoring political and demographic input, and instead simply optimize for compactness and other related metrics. However, we maintain that this is a flawed approach because compactness and fairness are orthogonal qualities; to achieve a meaningful notion of fairness, one needs to model political and demographic considerations, using historical data. We will discuss a series of papers that explore and develop this perspective. We first present a scalable approach to explicitly optimize for arbitrary piecewise-linear definitions of fairness; this employs a stochastic hierarchical decomposition approach to produce an exponential number of distinct district plans that can be optimized via a standard set partitioning integer programming formulation. This enables a large-scale ensemble study of congressional districts, providing insights into the range of possible expected outcomes and the implications of this range on potential definitions of fairness. Further work extending this shows that many additional real-world constraints can be easily adapted in this framework (such as minimal county splits as was recently required in Alabama legislation in response to the US Supreme Court decision Milligan v. Alabama). In addition, one can adapt the same framework to heuristically optimize for other fairness-related objectives, such achieving a targeted number of majority minority districts (and in taking this approach, achieving stronger results than obtained by a prominent randomized local search approach known as “short bursts”).
We also show that our optimization infrastructure facilitates the study of the design of multi-member districts (MMDs) in which each district elects multiple representatives, potentially through a non-winner-takes-all voting rule (as was proposed in H.R. 4000 in an earlier session of Congress). We carry out large-scale analyses for the U.S. House of Representatives under MMDs with different social choice functions, under algorithmically generated maps optimized for either partisan benefit or proportionality. We find that with three-member districts using Single Transferable Vote, fairness-minded independent commissions can achieve proportional outcomes in every state (up to rounding), and this would significantly curtail the power of advantage-seeking partisans to gerrymander.
This is joint work with Wes Gurnee, Nikhil Garg, David Rothschild, Julia Allen, Cole Gaines, David Domanski, Rares-Stefan Bucsa, and Daniel Brous.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: David Shmoys is the Laibe/Acheson Professor and Director of the Center for Data Science for Enterprise & Society at Cornell University. He obtained his PhD in Computer Science from the University of California at Berkeley in 1984, and held postdoctoral positions at MSRI in Berkeley and Harvard University, and a faculty position at MIT before joining the faculty at Cornell University. He was Chair of the Cornell Provost’s “Radical Collaborations” Task Force on Data Science and was co-Chair of the Academic Planning Committee for Cornell Tech. His research has focused on the design and analysis of efficient algorithms for discrete optimization problems, with applications including scheduling, inventory theory, computational biology, computational sustainability, and data-driven decision-making in the sharing economy. His work has highlighted the central role that linear programming plays in the design of approximation algorithms for NP-hard problems. He was awarded the 2022 INFORMS Optimization Society Khachiyan Prize, the 2023 INFORMS Morse Lectureship, and the 2024 INFORMS Kimball Medal. His book (co-authored with David Williamson), The Design of Approximation Algorithms, was awarded the 2013 INFORMS Lanchester Prize and his work on bike-sharing (joint with Daniel Freund, Shane Henderson, and Eoin O’Mahony) was awarded the 2018 INFORMS Wagner Prize. David is a Fellow of the ACM, INFORMS, and SIAM, and was an NSF Presidential Young Investigator.
Host: CAIS
More Info: https://cais.usc.edu/events/usc-cais-seminar-with-dr-david-shmoys/
Location: Michelson Center for Convergent Bioscience (MCB) - 101
Audiences: Everyone Is Invited
Contact: Hailey Winetrobe Nadel, MPH, CHES
Event Link: https://cais.usc.edu/events/usc-cais-seminar-with-dr-david-shmoys/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Automatic Evaluation of Clinical Notes Generated from Doctor-Patient-Conversations
Mon, Nov 25, 2024 @ 11:00 AM - 11:50 AM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Mojtaba Elyaderani, Data Science Specialist - Solventum Corporation, Health Information Systems Business
Talk Title: Automatic Evaluation of Clinical Notes Generated from Doctor-Patient-Conversations
Abstract: Detailed detailed clinical documentation based on doctor-patient conversations is a necessary yet burdensome task for physicians and is often cited as one of the leading causes of physician burn-out. One way to reduce the documentation workload on physicians is to hire medical scribes, who assist in writing clinical notes. However, this option is costly and difficult to scale, putting it beyond the reach of many practitioners. This has led to the emergence of the ``ambient clinical documentation'' framework, where the conversation between doctor and patient is recorded and transcribed (with the patient's permission) and passed to a clinically trained Language Model (LM) which generates the corresponding note. Despite the recent improvements in their performance, modern LMs still make many errors that are unacceptable in a medical setting and can generate clinical notes that are of poor quality. For example, they may miss critical information, contain hallucinated content, or include important information in wrong note sections. Delivering poor-quality notes to physicians can be an extra burden, potentially resulting in a more time-consuming note creation process than simply starting from scratch. This proves the necessity of evaluating LM-generated clinical notes in a scalable and efficient manner. In this presentation we will introduce few such approaches and examine their weaknesses and strengths.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Zoom Link: https://usc.zoom.us/j/95154565194?pwd=LMaaRHXgKCeabJ7UxufaOW3HUu5Ys2.1
Host: Associate Prof. Meisam Razaviyayn
Webcast: https://usc.zoom.us/j/95154565194?pwd=LMaaRHXgKCeabJ7UxufaOW3HUu5Ys2.1Location: Olin Hall of Engineering (OHE) - 136
WebCast Link: https://usc.zoom.us/j/95154565194?pwd=LMaaRHXgKCeabJ7UxufaOW3HUu5Ys2.1
Audiences: Everyone Is Invited
Contact: Thomas Lord Department of Computer Science
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
