Robert West

Robert West is a tenure-track assistant professor of computer science at EPFL, where he heads the Data Science Lab. In his research, he develops and applies techniques in machine learning, computational social science, natural language processing, social network analysis, and data mining. Bob also collaborates closely with the Wikimedia Foundation, in his role as a Wikimedia Research Fellow. Bob’s work has won several awards, including best/outstanding paper awards at ICWSM’21, ICWSM’19, and WWW’13, a best-paper runner-up award at WWW’16, a Google Faculty Research Award, a Facebook Research Award, a Hewlett-Packard Graduate Fellowship, and a Facebook Graduate Fellowship. He is actively involved in the research community, e.g., as an Associate Editor of ICWSM and EPJ Data Science and as a co-founder of the Wiki Workshop (held at WWW and ICWSM) and the Applied Machine Learning Days. Bob received his PhD in Computer Science from Stanford University, his MSc from McGill University, Canada, and his undergraduate degree from Technische Universität München, Germany.[Last updated: 25 Aug 2021]

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Unités associées (5)

Unités associées

Chargement

Cours enseignés par cette personne (3)

Cours enseignés par cette personne

Chargement

Domaines de recherche associés (3)

Domaines de recherche associés

Chargement

Publications associées (32)

Publications associées

Chargement

Personnes menant des recherches similaires (102)

Personnes menant des recherches similaires

Chargement

CS-401: Applied data analysis

This course teaches the basic techniques, methodologies, and practical skills required to draw meaningful insights from a variety of data, with the help of the most acclaimed software tools in the data science world (pandas, scikit-learn, Spark, etc.)

CS-727: Topics in Computational Social Science (TopiCSS)

This is a seminar course. By reading and discussing an introductory book as well as research papers about computational social science, students will become familiar with core issues and techniques in the field.

ENG-704: EECS Seminar: Advanced Topics in Machine Learning

Students learn about advanced topics in machine learning, artificial intelligence, optimization, and data science. Students also learn to interact with scientific work, analyze and understand strengths and weaknesses of scientific arguments of both theoretical and experimental results.

Wikipédia

Wikipédia () est une encyclopédie collaborative, généraliste et multilingue créée par Jimmy Wales et Larry Sanger le . Il s'agit d'une œuvre libre, c'est-à-dire que chacun est libre de la rediffuser.

Information

vignette|redresse=0.6|Pictogramme représentant une information. L’information est un de la discipline des sciences de l'information et de la communication (SIC). Au sens étymologique, l'« informatio

Technologie de l'information

La technologie de l'information (TI), ou IT pour « information technology » en anglais, appelée aussi système informatique, désigne le domaine technique du traitement de l'information, souvent dans u

A Large-Scale Characterization of How Readers Browse Wikipedia

Akhil Arora, Tiziano Piccardi, Robert West

Despite the importance and pervasiveness of Wikipedia as one of the largest platforms for open knowledge, surprisingly little is known about how people navigate its content when seeking information. To bridge this gap, we present the first systematic large-scale analysis of how readers browse Wikipedia. Using billions of page requests from Wikipedia's server logs, we measure how readers reach articles, how they transition between articles, and how these patterns combine into more complex navigation paths. We find that navigation behavior is characterized by highly diverse structures. Although most navigation paths are shallow, comprising a single pageload, there is much variety, and the depth and shape of paths vary systematically with topic, device type, and time of day. We show that Wikipedia navigation paths commonly mesh with external pages as part of a larger online ecosystem, and we describe how naturally occurring navigation paths are distinct from targeted navigation in lab-based settings. Our results further suggest that navigation is abandoned when readers reach low-quality pages. Taken together, these insights contribute to a more systematic understanding of readers' information needs and allow for improving their experience on Wikipedia and the Web in general.

ASSOC COMPUTING MACHINERY2023

Chargement

Language Model Decoding as Likelihood–Utility Alignment

Boi Faltings, Valentin Hartmann, Martin Josifoski, Maxime Jean Julien Peyrard, Frano Rajic, Jiheng Wei, Robert West

A critical component of a successful language generation pipeline is the decoding algorithm. However, the general principles that should guide the choice of a decoding algorithm re- main unclear. Previous works only compare decoding algorithms in narrow scenarios, and their findings do not generalize across tasks. We argue that the misalignment between the model’s likelihood and the task-specific notion of utility is the key factor to understanding the effectiveness of decoding algorithms. To struc- ture the discussion, we introduce a taxonomy of misalignment mitigation strategies (MMSs), providing a unifying view of decoding as a tool for alignment. The MMS taxonomy groups decoding algorithms based on their implicit assumptions about likelihood–utility misalign- ment, yielding general statements about their applicability across tasks. Specifically, by an- alyzing the correlation between the likelihood and the utility of predictions across a diverse set of tasks, we provide empirical evidence supporting the proposed taxonomy and a set of principles to structure reasoning when choos- ing a decoding algorithm. Crucially, our analy- sis is the first to relate likelihood-based decod- ing algorithms with algorithms that rely on ex- ternal information, such as value-guided meth- ods and prompting, and covers the most di- verse set of tasks to date. Code, data, and models are available at https://github.com/epfl- dlab/understanding-decoding.

2023

Chargement

Linguistic effects on news headline success: Evidence from thousands of online field experiments (Registered Report)

Kristina Gligoric, Robert West

What makes written text appealing? In this registered report, we study the linguistic characteristics of news headline success using a large-scale dataset of field experiments (A/B tests) conducted on the popular website comparing multiple headline variants for the same news articles. This unique setup allows us to control for factors that could otherwise have important confounding effects on headline success. Based on the prior literature and an exploratory portion of the data, we formulated hypotheses about the linguistic features associated with statistically superior headlines, previously published as a registered report protocol. Here, we report the findings based on a much larger portion of the data that became available after the publication of our registered report protocol. Our registered findings contribute to resolving competing hypotheses about the linguistic features that affect the success of text and provide avenues for research into the psychological mechanisms that are activated by those features.

PUBLIC LIBRARY SCIENCE2023

Chargement