AI & Fundamentals
Using Algorithms to Understand Transformers (and Using Transformers to Understand Algorithms) - Vatsal Sharan, Assistant Professor, University of Southern California
Using Algorithms to Understand Transformers (and Using Transformers to Understand Algorithms) - Vatsal Sharan, Assistant Professor, University of Southern California
Abstract:We will discuss how algorithmic tools and understanding borrowed from optimization theory, Fourier transforms, and Boolean function analysis can help understand the mechanisms employed by Transformers to solve basic computationa...
AI & Fundamentals
Recent Development of Generative AI Focused on Computer Vision - Dongjun Kim, Postdoctoral Scholar, Stanford University
Recent Development of Generative AI Focused on Computer Vision - Dongjun Kim, Postdoctoral Scholar, Stanford University
Abstract:With the recent advancements in diffusion models, the image generation problem has become nearly resolved in terms of quality. However, because most large language models (LLMs) are built on autoregressive transformer architectu...
AI & Fundamentals
Stochastic Approximation Algorithms with Decision-Dependent Data: The Case of Performative Prediction - Hoi-To Wai, Assistant Professor, CUHK
Stochastic Approximation Algorithms with Decision-Dependent Data: The Case of Performative Prediction - Hoi-To Wai, Assistant Professor, CUHK
Abstract:Stochastic approximation (SA) forms the foundation of numerous online decision-making algorithms under uncertainty. In recent years, it has garnered renewed interest in the dynamic environment setting where streaming data is not...
AI & Fundamentals
Tuning Free (inference time) Alignment of Large Language Models - Amrit Singh Bedi, Assistant Professor, University of Central Florida
Tuning Free (inference time) Alignment of Large Language Models - Amrit Singh Bedi, Assistant Professor, University of Central Florida
Abstract:Traditional fine-tuning of foundation models is computationally heavy, involving updates to billions of parameters. A promising alternative, alignment via decoding, adjusts the response distribution directly without model update...
AI & Fundamentals
LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language - James Requeima, Postdoctoral Fellow, University of Toronto
LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language - James Requeima, Postdoctoral Fellow, University of Toronto
Abstract:Machine learning practitioners often face significant challenges in formally integrating their prior knowledge and beliefs into predictive models, limiting the potential for nuanced and context-aware analyses. Moreover, the expe...
AI & Fundamentals
Efficient Fine-tuning via Model Reprogramming - Feng Liu, Assistant Professor, University of Melbourne
Efficient Fine-tuning via Model Reprogramming - Feng Liu, Assistant Professor, University of Melbourne
Abstract:Knowledge shouldn’t be limited to those who can pay," said Robert C. May, chair of UC's Academic Senate. In machine learning, this is particularly relevant, as recent foundation models—pre-trained on massive datasets—have widene...
AI & Fundamentals
Generative World Modeling for Embodied Agents - Sherry Yang, Assistant Professor, NYU
Generative World Modeling for Embodied Agents - Sherry Yang, Assistant Professor, NYU
Abstract:Generative models have transformed content creation, and the next frontier may be simulating realistic experiences in response to actions by humans and agents. In this talk, I will talk about a line of work that involves learnin...
AI & Fundamentals
Advancing Multimodal Vision-Language Learning - Aishwarya Agrawal, Assistant Professor, University of Montreal
Advancing Multimodal Vision-Language Learning - Aishwarya Agrawal, Assistant Professor, University of Montreal
Abstract:Over the last decade, multimodal vision-language (VL) research has seen impressive progress. We can now automatically caption images in natural language, answer natural language questions about images, retrieve images using comp...
AI & Fundamentals
Contrastive Learning for ML-guided MIP search - Bistra Dilkina, Associate Professor, University of Southern California
Contrastive Learning for ML-guided MIP search - Bistra Dilkina, Associate Professor, University of Southern California
Abstract:Recent research has demonstrated the ability to significantly improve MIP solving by integrating ML-guided components and learning policies tailored to specific problem distributions. This benefits directly the real-world deploy...
AI & Fundamentals
Highly Accurate Predictions on Small Data with the Tabular Foundation Model TabPFN - Frank Hutter, Professor, University of Freiburg
Highly Accurate Predictions on Small Data with the Tabular Foundation Model TabPFN - Frank Hutter, Professor, University of Freiburg
Abstract:Tabular data, spreadsheets organized in rows and columns, is ubiquitous in many fields, prominently including medicine. The fundamental prediction task of filling in a label column based on the rest of the columns (the so-called...