CDS Spring 2023, 2021 Colloquium
The CDS Colloquium Series was developed to build intellectual community within and beyond the academic unit. Since its inception, the series has welcomed dozens of scholars and is intended for CDS faculty, staff, and students. However, we welcome interest from across the Boston University campus and beyond.
To view all upcoming lectures, events, and programs, visit the CDS Calendar.
Past Colloquia
February 22, 2023
Socially Responsible & Factual Reasoning for Equitable AI Systems
Speaker: Saadia Gabriel, PhD candidate, Paul G. Allen School of Computer Science & Engineering, University of Washington, advised by Prof. Yejin Choi and Prof. Franziska Roesner
Understanding the implications underlying a text is critical to assessing its impact. This requires endowing artificial intelligence (AI) systems with pragmatic reasoning, for example to infer that the statement “Epidemics and cases of disease in the 21st century are “staged”” relates to unfounded conspiracy theories.
This presentation will outline how shortcomings in the ability of current AI systems to reason about pragmatics leads to inequitable detection of false or harmful language, and demonstrate how these shortcomings can be addressed by imposing human-interpretable structure on deep learning architectures using insights from linguistics.
February 16, 2023
Empowering Graph Neural Networks from a Data-Centric View
Speaker: Wei Jin, PhD candidate, Michigan State University, advised by Prof. Jiliang Tang
This presentation will provide a fresh perspective on enhancing graph inputs, graph neural networks (GNNs) by optimizing the graph data, rather than designing new models. Specifically, Jin will present a model-agnostic framework which improves prediction performance by enhancing the quality of an imperfect input graph. Then show how to significantly reduce the size of a graph dataset while preserving sufficient information for GNN training.
January 30, 2023
Towards Secure and Regulated Machine Learning Systems
Speaker: Emily Wenger, A fifth-year Ph.D. candidate in the SAND Lab at the University of Chicago, advised by Ben Zhao and Heather Zheng.
This presentation highlighted two key areas of Wegner’s work: vulnerabilities in and caused by ML models and a novel attack discovered against computer vision models. Wegner explored building practical tools that protect models and empower users, highlighted a privacy tool she developed to disrupt unwanted facial recognition, followed by discussion of her vision for the future of secure and regulated ML.
April 7, 2021
Democra-CS: Computational Perspectives on Democracy
Speaker: Anson Kahng
Kahng will present research on the theoretical foundations of two new paradigms of democracy: virtual democracy and liquid democracy.
April 1, 2021
Cryptography and the Democratizing Power of Learning Nothing
Speaker: Mayank Varia, BU
Varia will present research on cryptographically protected computing and its social, legal, and public policy impacts.
March 29, 2021
Reading and Writing the Rules Learned by Deep Networks
Speaker: David Bau, PhD student at MIT
Can a person understand how an AI works well enough to reprogram an AI directly? Bau will discuss how deep networks can be dissected to reveal the organization of their internal computations.
March 25, 2021
The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks
Speaker: Jonathan Frankle, fifth year PhD student at MIT
Frankle will dive into his recently proposed lottery ticket hypothesis: that the dense neural networks we typically train have much smaller subnetworks capable of reaching full accuracy from early in training.
March 24, 2021
Advancing Public Policy with Machine Learning
Speaker: Sabina Tomkins, postdoctoral scholar at Stanford Computational Policy Lab
Tomkins will present her work showing that environmental events can impact human trafficking activity. She will also discuss her work in digital health and present a state-of-the-art algorithm for learning the contexts under which to nudge users to be physically active.
March 22, 2021
Machine Learning and Causality: Building Efficient, Reliable Models for Decision-Making
Speaker: Maggie Makar, PhD student at CSAIL, MIT
Current techniques for causal inference typically rely on having access to large amounts of data, limiting their applicability to data-constrained settings. In addition, evidence has shown that most predictive models are insufficiently robust with respect to shifts at test time. Makar will present her work on building novel techniques addressing these problems.
March 16, 2021
Data-Driven Transfer of Insight between Brains and AI Systems
Speaker: Mariya Toneva, Ph.D. candidate in a joint program between Machine Learning and Neural Computation at Carnegie Mellon University
Toneva will present her research that establishes a direct connection between the human brain and AI systems with two main goals: 1) to improve the generalization performance of AI systems and 2) to improve our mechanistic understanding of cognitive functions.
March 15, 2021
Beyond Function Approximation: Compression, Inference, and Generation via Supervised Learning
Speaker: Jiaming Song, Ph.D. candidate in the Computer Science Department at Stanford University
Song proposes techniques to estimate and optimize divergences more effectively by leveraging advances in supervised learning
March 10, 2021
What Your AI Does Not Know
Speaker: Christos Tzamos, Assistant Professor in the Department of Computer Sciences at University of Wisconsin-Madison
Tzamos will show how to adapt influential ideas from statistics to create efficient algorithms that deal with inputs that are biased or unreliable
March 8, 2021
The Societal Impacts of Algorithmic Decision-Making
Speaker: Manish Raghavan, PhD candidate in the Computer Science department at Cornell University
Raghavan will discuss his efforts to develop principles for the responsible development and deployment of algorithmic decision-making systems.
March 4, 2021
Ideal Made Real: Machine Learning with Limited Data and Interpretable Outputs.
Speaker: David Alvarez-Melis, postdoctoral researcher in the Machine Learning and Statistics Group at Microsoft Research, New England
Alvarez-Melis will present various approaches that he has developed for 'amplifying' (e.g, merging, transforming, interpolating) datasets based on the theory of Optimal Transport.
March 1, 2021
Mechanism Design for Social Good
Speaker: Kira Goldner, postdoctoral researcher in the Computer Science Department and at the Data Science Institute at Columbia University
Golder will show how results from the theoretical foundations of algorithmic mechanism design can be used to solve problems of societal concern, focusing on applications in carbon license allocations, health insurance markets, and online labor markets.