Abstract: Trustworthy AI deployment in high-stakes domains requires systems that are fair, private, robust, and controllable as they scale. Yet these demands are often pursued through ad-hoc approaches, lacking a systematic understanding of the inherent trade-offs between competing objectives. We add fairness regularizers and hope bias decreases. We train on massive datasets and hope the model learns the underlying logic of how concepts combine, rather than memorizing statistical shortcuts. We encrypt data and hope the resulting computational overhead remains manageable. But hope isnot a science.
In this talk, I argue that what trustworthy AI lacks is not better heuristics but a deeper science of what these properties fundamentally cost and what is achievable. Before we can fix a system, we must map the terrain: what trade-offs are unavoidable, what regions of performance areunreachable, and how far current methods fall from what is actually achievable. My research builds this map across fairness, privacy, robustness, and controllability, following a common methodology: diagnose where models fail, characterize the fundamental limits any method must obey, and design systems that approach those limits. I will present this framework, its extension to scientific applications where we replace statistical constraints with physical laws to ensure AI systems remain grounded in reality, and a vision for scaling these principles to the rapidly expanding ecosystem of composed and interacting AI systems.


Bio: Dr. Vishnu Boddeti is an Associate Professor in the Department of Computer Science and Engineering at Michigan State University, where he leads the Human Analysis Lab (HAL). His research develops mathematical frameworks for trustworthy AI, spanning fairness, privacy, robustness, and physics-informed learning, with an emphasis on characterizing fundamental limits and building systems that achieve them. His work has been supported by NSF, NIST, DARPA, ONR, Ford, and others, and recognized with a Meta Research Award (2021). His research has been featured on the cover of Nature, recognized as an Editor's Highlight in Nature Communications, and received multiple best paper awards, including the 2024 IEEE-CCF Cloud Computing Best Paper Award and the TMLR Outstanding Certification Finalist (2023). He serves as Senior Area Editor for IEEE Transactions on Information Forensics and Security and completed his PhD in ECE from Carnegie Mellon University in 2012.

Location: NCS 120
Abstract: Many foundation models for digital pathology have been released recently. Benchmarking available methods then becomes paramount to get a clearer view of the research landscape. For this reason, we introduce THUNDER, a tile-level benchmark for digital pathology foundation models, allowing for efficient comparison of many models on diverse datasets with a series of downstream tasks, studying their feature spaces and assessing the robustness and uncertainty of predictions informed by their embeddings. Such foundation models are often used as feature extractors and combined with Multiple Instance Learning (MIL) aggregators at downstream time. Such aggregation must be efficient and reliable. We will focus on two specific examples of this: (I) HistAug, a fast and efficient generative model for controllable augmentations in the latent space of foundation models to perform data augmentation for MIL, and (ii) CAR-MIL, a method based on counterfactual attention regularisation to improve the reliability of attention maps of MIL methods.

Short-bio: Pierre Marza is a Postdoctoral Researcher at CentraleSupelec in the Biomathematics team of the MICS lab, studying Computer Vision and Deep Learning for Medical Imaging, with a focus on Digital Pathology. Prior to this, he was a PhD student at INSA Lyon, in the LIRIS and CITI labs, advised by Christian Wolf, and co-advised by Laetita Matignon and Olivier Simonin. He studied Visual Navigation, Embodied AI, Spatial Reasoning, more specifically how to learn to represent 3D space, generalize to new environments and master diverse tasks from light supervision.

Location: NCS 220

Zoom: https://stonybrook.zoom.us/j/94798224254?pwd=CFraer25qnpORbJ14aAVHRwaSJOjJM.1
CSE 600 Talk: Squeezing Software Performance via Eliminating Wasteful Operations presented by Xu Liu

ABSTRACT: Inefficiencies abound in complex, layered software. A variety of inefficiencies show up as wasteful memory operations, such as redundant or useless memory loads and stores. Aliasing, limited optimization scopes, and insensitivity to input and execution contexts act as severe deterrents to static program analysis. Microscopic observation of whole executions at instruction- and operand-level granularity breaks down abstractions and helps recognize redundancies that masquerade in complex programs. In this talk, I will describe various wasteful memory operations, which pervasively exist in modern
software packages and expose great potential for optimization. I will discuss the design of a fine-grained instrumentation-based profiling framework that identifies wasteful operations in their contexts, which guides nontrivial performance improvement. Furthermore, I will show our recent improvement to the profiling framework by abandoning
instrumentation, which reduces the runtime overhead from 10x to 3% on average. I will show how our approach works for native binaries and various managed languages such as Java, yielding new performance insights for optimization.

BIO: Xu Liu is an assistant professor in the Department of Computer Science at College of William & Mary. He obtained his PhD from Rice University in 2014 and joined the College of William & Mary in the same year. Prof. Liu works on building performance tools to pinpoint and optimize inefficiencies in HPC code bases. He has developed several open-source profiling tools, which are used worldwide at universities, DOE national laboratories and industrial companies. Prof. Liu has published a number of papers in high-quality venues. His papers received Best Paper Award at SC'15, PPoPP'18, PPoPP'19 and ASPLOS'17 Highlights, as well as Distinguished Paper Award at ICSE'19. His recent ASPLOS'18 paper has been selected as ACM SIGPLAN Research Highlights in 2019 and nominated for CACM Research Highlights. Prof. Liu is the receipt of 2019 IEEE TCHPC Early Career Researchers Award for Excellence in High Performance Computing. Prof. Liu served on the program committee of conferences such as SC, PPoPP, IPDPS, CGO, HPCA and ASPLOS.
University Libraries Presents: The Library AI Club is a welcoming space for students, faculty, and staff to explore AI in a supportive, low-pressure environment. Meeting every two weeks, the club features discussions, collaborative projects, guest speakers, and hands-on experiments. Join us to learn, share ideas, and engage with AI responsibly and creatively. We'd love to see you at an upcoming meeting! Location: Melville Library, Scholarly Communication Seminar Room
Title: Sustainable NLP

Time: Friday 4/29, 2:40 PM

Location: NCS 120

Abstract:


Natural language processing (NLP) technology has supercharged many real-world applications ranging from intelligent personal assistants (like Alexa, Siri, and Google Assistant) to commercial search engines such as Google and Bing. But current NLP applications use extremely large neural models, making them (i) expensive to deploy on servers, requiring large amounts of compute resources and power, and (ii) impossible to run on mobile devices, making on-device, privacy-preserving applications impractical.

In the first part of the talk, I will describe systems optimizations we have developed that significantly reduce the compute and memory requirement of NLP models. The optimizations we developed can be applied broadly and results in over 10x reduction in latency when deployed on mobile devices. In the second part of the talk, I will describe our recent work on predicting energy consumption of NLP models. Existing energy prediction approaches are not accurate, making it difficult for developers and practitioners to reason about their models in terms of power. We use a multi-level regression approach that produces highly accurate and interpretable energy predictions.



Bio:
Aruna Balasubramanian is an Associate Professor at Stony Brook University. She received her Ph.D from the University of Massachusetts Amherst, where her dissertation won the UMass outstanding dissertation award and was the SIGCOMM dissertation award runner up. She works in the area of networked systems. Her current work consists of two threads: (1) significantly improving Quality of Experience of Internet applications, and (2) improving the usability, accessibility, and privacy of mobile systems. She is the recipient of the SIGMobile Rockstar award, a Ubicomp best paper award, a Computing Innovation Fellowship, a VMWare Early Career award, several Google research awards, an


The International Conference on Learning Representations (ICLR) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence called representation learning, but generally referred to as deep learning.



ICLR is globally renowned for presenting and publishing cutting-edge research on all aspects of deep learning used in the fields of artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, text understanding, gaming, and robotics.

ICLR is one of the fastest growing artificial intelligence conferences in the world. Participants at ICLR span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.

The rapidly developing field of deep learning is concerned with questions surrounding how we can best learn meaningful and useful representations of data. ICLR takes a broad view of the field and includes topics such as feature learning, metric learning, compositional modeling, structured prediction, reinforcement learning, and issues regarding large-scale learning and non-convex optimization.

A non-exhaustive list of relevant topics explored at the conference include:



  • Unsupervised, Semi-supervised, and Supervised Representation Learning
  • Representation Learning for Planning and Reinforcement Learning
  • Metric Learning and Kernel Learning
  • Sparse Coding and Dimensionality Expansion
  • Hierarchical Models

  • Optimization for Representation Learning
  • Learning Representations of Outputs or States
  • Implementation Issues, Parallelization, Software Platforms, Hardware
  • Applications in Vision, Audio, Speech, Natural Language Processing, Robotics, Neuroscience, or Any Other Field


For more information or registration, please visit the official website.