<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Papers on Jannik P. Roth</title><link>https://jannik-roth.github.io/papers/</link><description>Recent content in Papers on Jannik P. Roth</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Fri, 12 Dec 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://jannik-roth.github.io/papers/index.xml" rel="self" type="application/rss+xml"/><item><title>Unraveling learning characteristics of transformer models for molecular design</title><link>https://jannik-roth.github.io/papers/2025_learning_characteristics/</link><pubDate>Fri, 12 Dec 2025 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2025_learning_characteristics/</guid><description>The learning characteristics of transformer based models for generative compound designs are studied using control calculations and careful manipulation of datasets. Published in Patterns, 2025</description></item><item><title>Protocol to calculate and compare exact Shapley values for different kernels in support vector machine models using binary features</title><link>https://jannik-roth.github.io/papers/2024_star/</link><pubDate>Fri, 20 Dec 2024 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2024_star/</guid><description>A protocol to calculate and compare exact Shapley values for support vector machine models with commonly used kernels and binary input features is developed. Published in STAR Protocols, 2024</description></item><item><title>Machine learning models with distinct Shapley value explanations decouple feature attribution and interpretation for chemical compound predictions</title><link>https://jannik-roth.github.io/papers/2024_exact_shapley/</link><pubDate>Wed, 21 Aug 2024 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2024_exact_shapley/</guid><description>A method for the exact calculation of Shapley Values for Support Vector Machines is introduced and tested for compound prediction tasks. Published in Cell Reports Physical Science, 2024</description></item><item><title>Relationship between prediction accuracy and uncertainty in compound potency prediction using deep neural networks and control models</title><link>https://jannik-roth.github.io/papers/2024_uq/</link><pubDate>Tue, 19 Mar 2024 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2024_uq/</guid><description>The prediction accuracy and uncertainty qunatification of deep neural networks and other control methods is compared using compound potency prediction tasks. Published in Scientific Reports, 2024</description></item><item><title>Chemical Reactivity of Supported ZnO Clusters: Undercoordinated Zinc and Oxygen Atoms as Active Sites</title><link>https://jannik-roth.github.io/papers/2020_zno/</link><pubDate>Wed, 28 Oct 2020 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2020_zno/</guid><description>The growth of ZnO clusters supported by ZnO-bilayers on Ag(111) and the interaction of these oxide nanostructures with water have been studied by a multi-technique approach combining temperature-dependent infrared reflection absorption spectroscopy (IRRAS), grazing-emission X-ray photoelectron spectroscopy, and density functional theory calculations. Published in ChemPhysChem, 2020</description></item><item><title>Influence of Strain on Acid–Basic Properties of Oxide Surfaces</title><link>https://jannik-roth.github.io/papers/2020_strain/</link><pubDate>Wed, 12 Aug 2020 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2020_strain/</guid><description>This paper studies the influence of strain on the acid-base properties of oxide surfaces using density functional theory. Published in The Journal of Physical Chemistry C, 2020</description></item><item><title>Precursor chemistry of h-BN: adsorption, desorption, and decomposition of borazine on Pt(110)</title><link>https://jannik-roth.github.io/papers/2020_borazine/</link><pubDate>Mon, 04 May 2020 00:00:00 +0000</pubDate><guid>https://jannik-roth.github.io/papers/2020_borazine/</guid><description>This paper studies the adsorption, desorption and fragmentation of borazine on Pt(110) by temperature-programmed desorption, ultraviolet photoemission spectroscopy, workfunction measurements and density functional theory. Published in Physical Chemistry Chemical Physics, 2020</description></item></channel></rss>