Wednesday / CS Seminar
January 22nd / 12:00 PM
Hybrid / Mudd 3514

Speaker
Ankit Garg, Microsoft Research India

Talk Title
Arithmetic circuits: lower bounds, learning and applications

Abstract
Arithmetic circuits are a natural model for computing polynomials via the basic operations of addition and multiplication. One of the fundamental open problems in this area is of proving lower bounds, i.e. finding an explicit polynomial that cannot be computed by polynomial sized arithmetic circuits (aka the VP vs VNP problem). While the question of proving lower bounds for general arithmetic circuits is still open, there has been remarkable progress in proving lower bounds for restricted classes of arithmetic circuits. Another important problem in this area is that of learning arithmetic circuits: given a polynomial (via query or black box access), output a small arithmetic circuit computing it (if one exists). This problem is hard in the worst case. I will present a meta framework for learning arithmetic circuits in the non-degenerate case using lower bound methods. We instantiate and implement this meta framework for various classes of arithmetic circuits. Then I will talk about extending the algorithms to the noisy setting as well as surprising and remarkable applications to classical problems in machine learning such as subspace clustering and mixtures of Gaussians. This is based on joint works with Pritam Chandra, Neeraj Kayal, Kunal Mittal, Chandan Saha and Tanmay Sinha.

Biography
Ankit Garg is a Senior Researcher at Microsoft Research India since July 2018. Prior to this, he was a Postdoctoral Researcher at Microsoft Research New England from 2016 - 2018. He completed his PhD in 2016 from Princeton University under the supervision of Prof Mark Braverman. His research has spanned several areas of theoretical computer science such as communication complexity, arithmetic complexity, optimization, mixture models and optimization. Several of his research papers have been published in top conferences in computer science such as STOC, FOCS, CCC, NeuRIPS and QIP, and recognized by Simons award for graduate students in theoretical computer science and a Siebel scholarship.

Research/Interest Areas
Theoretical Computer Science, Algorithms, Computational Complexity Theory

---
Zoom: https://northwestern.zoom.us/j/92525649422?pwd=N6DlLBx6RRAtrar8vk27te6cjmzqXP.1
Panopto: https://northwestern.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=4d9fd60c-89e5-48f6-bbdb-b261015638de
DEI Minute: tinyurl.com/cspac-dei-minute

Abstract: For centuries, researchers have sought out ways to connect disparate areas of knowledge. With the advent of Artificial Intelligence (AI), we can now rigorously explore relationships that span across distinct areas – such as, mechanics and biology, or science and art – to deepen our understanding, to accelerate innovation, and to drive scientific discovery. However, many existing AI methods have limitations when it comes to physical intuition, and often hallucinate. To address these challenges, we present research that blurs the boundary between physics-based and data-driven modeling through a series of physics-inspired multimodal graph-based generative AI models, set forth in a hierarchical multi-agent mixture-of-experts framework. The design of these models follows a biologically inspired approach where we re-use neural structures and dynamically arrange them in different patterns and utility, implementing a manifestation of the universality-diversity-principle that forms a powerful principle in bioinspired materials. This new generation of models is applied to the analysis and design of materials, specifically to mimic and improve upon biological materials. Applied specifically to protein engineering, the talk will cover case studies covering distinct scales, from silk, to collagen, to biomineralized materials, as well as applications to medicine, food and agriculture where materials design is critical to achieve performance targets.

Bio: Markus J. Buehler is the McAfee Professor of Engineering at MIT. Professor Buehler pursues new modeling, design and manufacturing approaches for advanced bio-inspired materials that offer greater resilience and a wide range of controllable properties from the nano- to the macroscale. He received many distinguished awards, including the Feynman Prize, the ASME Drucker Medal, the J.R. Rice Medal, and many others. Buehler is a member of the National Academy of Engineering.

Related Info

Join Machine Learning and Data Science (MLDS) and McCormick as we host a panel of MLDS alumni to speak more about their experience postgrad working in various industries from 2-3pm on January 22nd. Our panelists include alumni working at AlixPartners, Abbott Diagnostics, Marsh, Ryder System, United Airlines, and Uber Technologies in roles such as Software Engineering, Industrial Engineering Analysis, and Consulting. This event will be on Zoom so be sure to join at 1:45pm to ensure you’re able to hear from all the members of our panel. All students regardless of major, minor, or school are welcome to join this event so feel free to tell any friends who may be interested. Join at this link (or paste https://northwestern.zoom.us/j/95508921001 into your browser) and if you have any difficulties feel free to email dseadmin@u.northwestern.edu. We look forward to seeing you there!

Members of the NITMB community are invited to join us for Research-In-Progress meetings, an informal venue for members of the NITMB to discuss ongoing and/or planned research. Efe Gökmen is an NITMB Fellow. Gökmen’s expertise lies at the crossroads of machine learning, statistical physics, and information theory. Learn more about Efe Gökmen’s research and engage in discussion with the NITMB community. Research-In-Progress talks take place on Wednesdays at 3pm at the NITMB office (875 N Michigan Ave., Suite 4010). Snacks and coffee will follow.

Related Info

Are you looking to teach a class where your students need access to GPUs? Do you need more GPUs than are available on Quest to train new AI models? Do you have specific hardware needs or are you hoping to deploy a science gateway? ACCESS-CI and NAIRR (National AI Research Resource) are two federally-funded computing resources available to support these and other computing needs. This workshop will introduce you to these platforms and cover how to set up accounts and request resources. This workshop will not be recorded.

Prerequisites: It may be helpful to have an account on the ACCESS-CI platform. You can sign up for an account on the ACCESS CI Registration Page if interested.

Register Now

Abstract:

Despite decades of research, signalized intersections remain a major urban bottleneck and traffic signal timing in practice is suboptimal. Signal timing algorithms must address two challenges: performance under uncertainty in future demand and turning proportions, and real-time computation. One possible approach is max-pressure signal timing. By modeling the traffic network as a Markov decision process, max-pressure control is mathematically proven to maximize throughput under uncertainty using Lyapunov drift. Nevertheless, the control itself is easy to compute with the technical difficulty relegated to the mathematical analysis of throughput properties. Recent work on max-pressure signal timing has integrated some practicalities of traffic signal timing into the mathematical control and analyses, such as cyclical phase selection, pedestrian phases, signal coordination, transit signal priority, and limited deployment. Moreover, simulation results comparing max-pressure control against current signal timings in Hennepin County corridors suggest significant improvements from using max-pressure control. This seminar will introduce max-pressure control and then present recent work on bridging the mathematical theory with the practice of signal timing towards implementation on public roads.

Bio:

Michael W. Levin is an Associate Professor in the Department of Civil, Environmental, and Geo- Engineering at the University of Minnesota. He received a B.S. degree in Computer Science and a Ph.D. degree in Civil Engineering from The University of Texas at Austin in 2013 and 2017, respectively. Dr. Levin is a member of the Network Modeling Committee (AEP40) of the Transportation Research Board and is on the editorial board of Transportation Research Part B: Methodological. His work has been published in top journals including Transportation Science, Transportation Research Part C: Emerging Technologies, and IEEE Transactions on Intelligent Transportation Systems and has received several awards, including the 2019 Ryuichi Kitamura Award and the 2016 Milton Pikarsky Award from the Council of University Transportation Centers. His research focuses on traffic flow and network modeling of connected autonomous vehicles and intelligent transportation systems.

The Role of AI in Scientific Discovery: Opportunities and Limitations

Xiangliang Zhang, Leonard C. Bettex Collegiate Professor of Computer Science, University of Notre Dame

Abstract: Artificial Intelligence (AI) is reshaping the landscape of scientific discovery, enabling breakthroughs across diverse fields. However, when these AI tools are applied to scientific problems, gaps and mismatches often arise. The inherent uncertainty in scientific phenomena, coupled with issues like data quality, biases, and interpretability, poses significant challenges. This talk will discuss the transformative potential of AI in scientific discovery, focusing on its applications in predictive modeling, generative tasks, optimization strategies, and literature analysis. Examples will include AI models ranging from traditional neural networks to large language models (LLMs). At the same time, their limitations will be critically examined, calling for collaboration between the AI and scientific communities to address these challenges and unlock AI’s full potential in advancing scientific discovery.

Related Info

Friday / CS Seminar
January 24th / 12:00 PM
Hybrid / Mudd 3514

Speaker
Meena Jagadeesan, UC Berkeley

Talk Title
Steering Machine Learning Ecosystems of Interacting Agents

Abstract

"Modern machine learning models—such as LLMs and recommender systems—interact with humans, companies, and other models in a broader ecosystem. However, these multi-agent interactions often induce unintended ecosystem-level outcomes such as clickbait in classical content recommendation ecosystems, and more recently, safety violations and market concentration in nascent LLM ecosystems.

In this talk, I discuss my research on characterizing and steering ecosystem-level outcomes. I take an economic and statistical perspective on ML ecosystems, tracing outcomes back to the incentives of interacting agents and to the ML pipeline for training models. First, in LLM ecosystems, we show how analyzing a single model in isolation fails to capture ecosystem-level performance trends: for example, training a model with more resources can counterintuitively hurt ecosystem-level performance. To help steer ecosystem-level outcomes, we develop technical tools to assess how proposed policy interventions affect market entry, safety compliance, and user welfare. Then, turning to content recommendation ecosystems, we characterize a feedback loop between the recommender system and content creators, which shapes the diversity and quality of the content supply. Finally, I present a broader vision of ML ecosystems where multi-agent interactions are steered towards the desired algorithmic, market, and societal outcomes."

Biography
Meena Jagadeesan is a 5th year PhD student in Computer Science at UC Berkeley, where she is advised by Michael I. Jordan and Jacob Steinhardt. Her research investigates multi-agent interactions in machine learning ecosystems from an economic and statistical perspective. She has received an Open Philanthropy AI Fellowship and a Paul and Daisy Soros Fellowship.

Research/Interest Areas
Artificial Intelligence, Machine Learning, Economics and Computation
---
Zoom: https://northwestern.zoom.us/j/92665903240?pwd=YGOjSzB0Pxk4oDvRsVBbA0auFIvH0p.1
Panopto: https://northwestern.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=408ccd84-8bf4-409a-b995-b261015654c8
DEI Minute: tinyurl.com/cspac-dei-minute

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Tensor Time Series: Factor Modeling and Deep Neural Networks

Yuefeng Han, Assistant Professor, Department of Applied and Computational Mathematics and Statistics, University of Notre Dame

Abstract: The analysis of tensors (multi-dimensional arrays) has become a vital area in modern statistics and data science, driven by advancements in scientific research and data collection. High-dimensional tensor data arise in diverse applications such as economics, genetics, microbiome studies, brain imaging, and hyperspectral imaging. These tensors are often high-dimensional and high-order, yet key information typically resides in reduced-dimensional subspaces governed by structural properties. This talk explores novel methodologies and theories for tensor time series analysis.

The presentation consists of two parts. The first part introduces a factor modeling framework for high-dimensional tensor time series, leveraging a structure similar to CP tensor decomposition. We propose a computationally efficient estimation procedure incorporating a warm-start initialization and an iterative simultaneous orthogonalization scheme. The algorithm achieves $\epsilon$-accuracy within $\log\log(1/\epsilon)$ iterations. Additionally, we establish inferential results, demonstrating consistency and asymptotic normality under relaxed assumptions. The second part integrates tensor factor models with deep neural networks. Specifically, a Tucker-type low-rank tensor structure is employed as a tensor-augmentation module in neural networks. Extensive experiments demonstrate the integration of this module into transformers and temporal neural networks for tensor time series prediction and tensor-on-tensor regression. The results highlight significant performance improvements, underscoring its potential for advancing time series forecasting.

Related Info

Friday / CS Seminar
January 31st / 12:00 PM
Hybrid / Mudd 3514

Speaker
Daniel Halpern, Harvard University

Talk Title
Why AI Needs Social Choice

Abstract
In many modern AI paradigms, we encounter tasks reminiscent of social choice theory: collecting preferences from individuals and aggregating them into a single joint outcome. However, these tasks differ from traditional frameworks in two key ways: the space of possible outcomes is so enormous that we can only hope to collect sparse inputs from each participant, and the outcomes themselves are often highly complex. This talk explores these challenges through two case studies: Polis, a platform for democratic deliberation (https://arxiv.org/abs/2211.15608), and Reinforcement Learning From Human Feedback (RLHF), a method for fine-tuning LLMs to align with societal preferences (https://arxiv.org/pdf/2405.14758). In both cases, the focus is on evaluating existing methods through an axiomatic lens and designing new methods with provable guarantees.

Biography
Daniel Halpern is a final-year PhD student at Harvard University advised by Ariel Procaccia. He is supported by an NSF Graduate Research Fellowship and a Siebel Scholarship. His research broadly sits at the intersection of algorithms, economics, and artificial intelligence. Specifically, he considers novel settings where groups of people need to make collective decisions, such as summarizing population views on large-scale opinion aggregation websites, using participant data to fine-tune large language models, and selecting panel members for citizens’ assemblies. In each, he develops practical and provably fair solutions to aggregate individual preferences.

---
Zoom: https://northwestern.zoom.us/j/96824870974?pwd=QfxpsRpfWcDlx4TXPswbAd8X4Dqhyb.1
Panopto: https://northwestern.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=2159520b-1d46-4ece-ba17-b268017475c6
DEI Minute: tinyurl.com/cspac-dei-minute

Would you like to simplify your machine learning code and minimize repetitive tasks? Scikit-Learn's pipelines can help you organize and streamline your data processing and model training, as well as make your code cleaner and easier to manage. In this workshop, we will cover why and how to use pipelines in your machine learning code.

Prerequisites: Participants should be familiar with Python at the level of the Python Fundamentals Bootcamp, another introductory Python workshop, or be a self-taught Python coder. Basic familiarity with machine learning and Scikit-Learn is required.

Register Now

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

"Using Computational Approaches to Understand the Social and Structural Drivers of Health"

By Michelle Birkett, Associate Professor of Medical Social Sciences (Determinants of Health) and Preventive Medicine and IPR Associate

This event is part of the Fay Lomax Cook Winter 2025 Colloquium Series, where our researchers from around the University share their latest policy-relevant research.

Please note all colloquia this quarter will be held in-person only.

In recent years generative artificial intelligence (GAI) tools have increased in availability and popularity at universities, but not everyone knows how to use them for better library search results. In this 60-minute hybrid session we will demonstrate how to use Microsoft Copilot, a free GAI tool for all NU students and faculty, to improve searches with the gold standard of reliable content, library databases. This session is geared toward participants with little-to-no familiarity with generative artificial intelligence, and will also address the basics of what generative artificial intelligence is, how it works, its risks, and limitations. Students can learn about NU’s Copilot accounts here.

This workshop is presented by Tracy Coyne, Distance Learning and Professional Studies Librarian; Frank Sweis, User Experience Librarian; and Jeannette Moss, User Education Librarian.

A Northwestern Zoom Account is required to access this session.

Register Now

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

AI for Nature: From Science to Impact

Tanya Berger-Wolf, Professor, Computer Science and Engineering and Director, Translational Data Analytics Institute, The Ohio State University

Abstract: Computation has fundamentally changed the way we study nature. New data collection technologies, such as GPS, high-definition cameras, autonomous vehicles under water, on the ground, and in the air, genotyping, acoustic sensors, and crowdsourcing, are generating data about life on the planet that are orders of magnitude richer than any previously collected. Yet, our ability to extract insight from these data lags substantially behind our ability to collect it.

The need for understanding is more urgent ever and the challenges are great. We are in the middle of the 6th extinction, losing the planet's biodiversity at an unprecedented rate and scale. In many cases, we do not even have the basic numbers of what species we are losing, which impacts our ability to understand biodiversity loss drivers, predict the impact on ecosystems, and implement policy.

The talk will discuss how AI can turn these data into high resolution information source about living organisms, enabling scientific inquiry, conservation, and policy decisions. It will introduce a new field of science, imageomics, and present a vision and examples of AI as a trustworthy partner both in science and biodiversity conservation, discussing opportunities and challenges.

Related Info

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Drawing on the Northwestern Engineering whole-brain philosophy and leadership in cognitive science, the Master of Science in Artificial Intelligence program would like to invite you to learn more at our upcoming webinar.

Take this opportunity to join Dr. Kristian Hammond, Professor of Computer Science and director of the MSAI program, as he discusses the complexities of this field, and how this newly offered program at Northwestern Engineering will prepare students for a career in artificial intelligence. At the end of the presentation, we will offer an open Q&A where you will be able to have your specific questions answered. You are also welcome to email your questions to us ahead of the session (msai@northwestern.edu).

Related Info

Register Now

Curious about how AI can transform your text data analysis? Large Language Models (LLMs), like those behind ChatGPT, offer powerful ways to extract information from text, such as identifying key individuals, finding information about events, or selecting sections of documents. Learn how LLMs can assist with information extraction, and how they compare with other approaches such as rule-based methods, regular expressions, and pre-trained entity recognition models.

Prerequisites: Open to anyone working with or interested in text data, this workshop will provide hands-on examples in Python, though the concepts apply across various programming languages. While prior experience in natural language processing is helpful, it’s not required to participate.

Register Now

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Towards Data-efficient Training of Large Language Models (LLMs)

Baharan Mirzasoleiman, Assistant Professor, Computer Science Department, UCLA

Abstract: High quality data is crucial for training LLMs with superior performance. In this talk, I will present two theoretically-rigorous approaches to find smaller subsets of examples that can improve the performance and efficiency of training LLMs. First, I will present a one-shot data selection method for supervised fine-tuning of LLMs. Then, I'll talk about an iterative data selection strategy to pretrain or fine-tune LLMs on imbalanced mixtures of language data. I'll conclude by showing empirical results confirming that the above data selection strategies can effectively improve the performance of various LLMs during fine-tuning and pretraining.

Related Info

Register Now

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute

Monday / CS Seminar
January 13th / 12:00 PM
Hybrid / Mudd 3514

Speaker
TBA

Talk Title
TBA

Abstract
TBA

Biography
TBA

Research/Interest Areas
TBA
---
Zoom: TBA
Panopto: TBA
DEI Minute: tinyurl.com/cspac-dei-minute