Do. 11.01.2024
Gastredner: Prof. Victor Veitch, University of Chicago
Präsentation: Linear Structure of High-Level Concepts in Text-Controlled Generative Models
Abstract: Text controlled generative models (such as large language models or text-to-image diffusion models) operate by embedding natural language into a vector representation, then using this representation to sample from the model's output space. This talk concerns how high-level semantics are encoded in the algebraic structure of representations. In particular, we look at the idea that such representations are ''linear''---what this means, why such structure emerges, and how it can be used for precision understanding and control of generative models.
Do. 18.01.2024
Gastredner: Prof. Jann Spiess, Stanford Graduate School of Business
Präsentation: Machine Learning Who to Nudge: Causal vs Predictive Targeting in a Field Experiment on Student Financial Aid Renewal
Do. 08.02.2024
Gastredner: Prof. Carlos Fernández-Loría, The Hong Kong University of Science and Technology
Präsentation: Causal Scoring: A Framework for Effect Estimation, Effect Ordering, and Effect Classification