Christoph Best
Christoph joined Google in 2010 from European Bioinformatics Institute, where he led the Electron Microscopy Data Bank project. Before that, he did research in Bioimage Informatics/Structural Biology at Max Planck Institute for Biochemistry in Munich, Germany, and Computational High Energy Physics at the John von Neumann Institute for Computation, Jülich, He did postdoctoral training in bioinformatics and theoretical physics at the University of Munich and the Massachusetts Institute of Technology, and holds a Dr. phil. nat. in Theoretical Physics from the University of Frankfurt.
Authored Publications
Sort By
Ars gratia retium: Understanding How Artificial Neural Networks Learn To Emulate Art
Algorithmic and Aesthetic Literacy: Emerging Transdisciplinary Explorations for the Digital Age, Verlag Barbara Budrich, Stauffenbergstr. 7
51379 Leverkusen
Germany (2021)
Preview abstract
Chapter of a book "Algorithmic and Aesthetic Literacy Matter" by Lydia Schulze Heuling (Bergen, Norway) and Christian Fink (Flensburg, Germany).
Discusses the process of generating art through artificial neural networks on a high level with an intended audience of teachers and teacher-training institutions. The context is the connection between algorithmic and aesthetic literacy. Contains a high-level introduction of artificial neural networks with some historical perspective, an exposition of the relationship of computer algorithms and aesthetic expression especially where used in teaching programming, and an overview of DeepDream and Generative Neural Networks with some examples. It ends with a discussion of uses of generative neural networks as artistic tools and their perspective in education.
View details
Advertising on YouTube and TV: A Meta-analysis of Optimal Media-mix Planning
Georg M. Goerg
Sheethal Shobowale
Jim Koehler
Nicolas Remy
Journal of Advertising Research (JAR), 57 (2015), pp. 283-304 (to appear)
Preview abstract
In this work we investigate under what circumstances a TV campaign should be complemented with online advertising to increase combined reach. First, we use probabilistic models to derive necessary and sufficient conditions. We then test these optimality conditions on empirical findings of a large collection of TV campaigns to answer two important questions: i) which characteristics of a TV campaign make it favorable to shift part of its budget to online advertising?; and ii) if it should shift, how much cost savings and additional reach can advertisers expect? First, we use classification methods such as linear discriminant analysis, logistic regression, and decision trees to decide whether a TV campaign should add online advertising; secondly, we train linear and support vector regression models to predict optimal budget allocation, cost savings, or additional reach. To train these models we use optimization results on roughly 26,000 campaigns. We do not only achieve excellent out-of-sample predictive power, but also obtain simple, interpretable, and actionable rules that improve the understanding of media mix advertising.
View details
Preview abstract
With vast increases in biological data generation, mechanisms for data storage and analysis have become limiting. A data structure, semantically typed data hypercubes (SDCubes), that combines hierarchical data format version 5 (HDF5) and extensible markup language (XML) file formats, now permits the flexible storage, annotation and retrieval of large and heterogenous datasets.
View details
