Data and Compute-Efficient Optimization of Parsimonious Models and Its Applications in Machine Learning and Payment Channel Networks
Dr. Christian Kümmerle University of North Carolina Monday, January 13, 2025 11:00AM – 12:00PM MSB 318 | Zoom Abstract For a machine learning model’s ability to generalize well to unseen data, it has been understood that it needs to be able to capture a hidden, low-dimensional data distribution in the high-dimensional feature space. In this…...
Read MoreCausal Machine Learning: Continuous structure learning and identifiability of causal invariances
Dr. Kevin Bello Soroco Thursday, November 21, 2024 11:00AM – 12:00PM HEC 101A | Zoom Abstract Interpretability and causality are key desiderata in modern machine learning systems. Graphical models, and more specifically directed acyclic graphs (DAGs, a.k.a. Bayesian networks), serve as a well-established tool for expressing interpretable causal relationships. However, the task of estimating DAG…...
Read MoreTowards Trustworthy AI: From Computer Vision to Medical Imaging
Dr. Yu Tian University of Pennsylvania Thursday, November 14, 2024 11:00AM – 12:00PM HEC 101A | Zoom Abstract In today’s world, deep neural networks drive machine learning systems that pervade every aspect of our daily lives. Yet, their deployment often raises concerns about trustworthiness, including safety, fairness, and other issues. This is particularly important and…...
Read MoreTowards Label Efficiency and Privacy Preservation in Video Understanding
Announcing the Final Examination of Ishan Dave for the degree of Doctor of Philosophy in Computer Science...
Read MoreTowards Trustworthy AI in Medical Image Analysis
Dr. Pingkun Yan Rensselaer Polytechnic University Tuesday, October 22, 2024 2:00PM – 2:45PM HEC 101B | Zoom Abstract Despite advancements in AI models for medical image analysis, ensuring their trustworthiness remains a challenge. Accuracy, interpretability, robustness, and generalizability are critical for gaining the confi-dence of healthcare professionals and other stakeholders. This talk focuses on AI…...
Read MoreTowards exponentially cheaper AI
Dr. Aditya Desai UC Berkeley Thursday, August 29, 2024 10:00AM – 11:00AM HEC 101A | Zoom Abstract The recent advancements in the capabilities of AI models have been extraordinary. However, training and deploying these models is prohibitively costly. The primary reason for increasing costs is the exponential increase in model sizes, which requires commensurate computing and…...
Read MoreStructuring Cooperative Teams for Multi-Agent Reinforcement Learning
Dr. Qi Zhang University of South Carolina Thursday, August 22, 2024 10:00AM – 11:00AM HEC 101A | Zoom Abstract Cooperative artificial intelligence equips a team of autonomous agents with the capability of planning and learning to maximize their joint utility, which finds a wide range of applications. Current solutions to cooperative AI, instantiated as cooperative…...
Read MoreResearch Experiences for Undergraduates (REU) Poster Session
Announcing the Final Examination of Matias Mendieta for the degree of Doctor of Philosophy in Computer Science...
Read MoreEfficient and Effective Dee Learning for Computer Vision in Centralized and Distributed Applications
Announcing the Final Examination of Matias Mendieta for the degree of Doctor of Philosophy in Computer Science...
Read MoreVideo Action Understanding: Advancing Action Recognition, Temporal Localization and Detection
Announcing the Final Examination of Praveen Tirupattur for the degree of Doctor of Philosophy in Computer Science...
Read MoreOn learning distributions: from dynamical system to generative modeling
Dr. Nicole Yang Emory University Tuesday, May 21, 2024 11:00AM – 12:00PM R1 101A | Zoom Abstract Dynamical systems are widely used to model complex real-world phenomena. They are models based on physical principles, which are highly interpretable but have limits in expressive capabilities. Furthermore, the already high dimensional, and complicated behavior make them difficult…...
Read MoreTowards Explainable and Reliable AI Models for Optimization
Dr. Jialin Liu DAMO Academy, Albaba Group US Thursday, April 25, 2024 10:00AM – 11:00AM TCII 222 | Zoom Abstract AI and data science have demonstrated remarkable potential in enhancing optimization algorithms. Compared with traditional methods, utilizing AI/ML techniques can potentially offer improvements in aspects like computational speed and solution quality. Despite these advancements, a…...
Read MoreAn Efficient One-Class SVM for Novelty Detection in IoT
Dr. Kun Yang Princeton University Monday, April 22, 2024 10:00AM – 11:00AM TCII 222 | Zoom Abstract Novelty detection is important in the Internet of Things (“IoT”) due to the potential threats that IoT devices can present. One-Class Support Vector Machines (OCSVMs) are one of the common approaches for novelty detection due to their ability…...
Read MoreBackdoors and bias in text-to-image generative models
Dr. Ajmal Mian The University of Western Australia Thursday, April 18, 2024 12:00PM – 1:00PM HEC 101 | Zoom Abstract In this presentation, I will explore the manipulation of text-to-image (T2I) generative models through backdoors. I will present our Backdoor Attack on Generative Models (BAGM), which infuses the generated images with subtle manipulative details by…...
Read MoreUnleashing the Power of Discrete Optimization in the New Era of AI
Dr. Arman Zharmagambetov Fundamental AI Research (FAIR) group at Meta Thursday, April 11, 2024 10:00AM – 11:00AM HEC 101A | Zoom Abstract Modern machine learning (ML) models, trained on real-world data, now underpin a broad spectrum of applications. Behind the success of these models, discrete optimization lays the foundation of the modeling, decision making and…...
Read MoreBridging the Gap: Translational AI in Biomedicine and Healthcare
Dr. Laura Brattain UCF College of Medicine, AI for Healthcare Monday, April 1, 2024 12:00PM – 1:00PM HEC 101 Abstract Recent research in biomedical AI has demonstrated the ability to not only aid in disease diagnosis and prognosis, but also facilitate procedure guidance and empower precision medicine approaches tailored to individual patients. To translate research…...
Read MoreQuantifying Uncertainties of Deep Neural Networks and Its Applications
Mr. Haomiao Ni Pennsylvania State University Thursday, March 14, 2024 10:30AM – 11:30AM HEC 101A | Zoom Abstract IIn the past two years, transformative AI products such as ChatGPT have underscored the profound impact of AI in our daily lives. Beyond natural language and image understanding, AI is revolutionizing healthcare, offering advantages such as assisting…...
Read MoreQuantifying Uncertainties of Deep Neural Networks and Its Applications
Dr. Jae Oh Woo Illinois Institute of Technology Tuesday, March 12, 2024 1:00PM – 2:00PM HEC 101A | Zoom Abstract In this presentation titled “Quantifying Uncertainties of Deep Neural Networks and Its Applications,” we delve into the critical task of measuring uncertainties in Deep Neural Networks. The talk will commence with a concise overview of…...
Read MoreAdversarial Graph Machine Learning on Blockchains
Dr. Cuneyt Akcora UCF College of Business, AI for Finance Monday, March 4, 2024 12:00PM – 1:00PM HEC 101 Abstract Blockchains allow pseudo-anonymous transactions, which has made it easier to create a payment ecosystem used worldwide. However, the ease of blockchain use has also attracted e-crime actors with malicious activities ranging from sextortion to ransomware…...
Read MoreOptimization for Fairness-aware Machine Learning
Ms. Yao Yao University of Iowa Thursday, February 8, 2024 11:00AM – 12:00PM TC 222 | Zoom Abstract Artificial intelligence (AI) and machine learning technologies have been used in high-stakes decision making systems such as lending decision, criminal justice sentencing and resource allocation. A new challenge arising with these AI systems is how to avoid…...
Read More